I. TOTAL SYNTHESIS OF DAN SHEN DITERPENOID QUINONES
II. SYNTHESIS AND CHEMISTRY OF (TRIALKYLSILYL)VINYLKETENES
by
JENNIFER LYNN LOEBACH
B.S., University of Illinois (1990)
SUBMITFTEDTO THE DEPARTMENT OF CHEMISTRY
IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY
June 1995
C Massachusetts Institute of Technology, 1995
Signature
of Author
..................................
Department of Chemistry
...............
May 5, 1995
Certified by...............
....... .........
Rick L. Danheiser
Thesis Supervisor
..
.........
I
,%1
Accepted
by.......... . ............................................
Dietmar Seyferth
Department Commitee on Graduate Studies
Science
MASSACHUSETTS
INSTITUTE
O Tfr-Nf0l 0GY
1
JUN 12 1995
LIBRARIES
This doctoral thesis has been examined by a committee of the Department of
Chemistry as follows:
/
,
Professor Peter T. Lansbury......
-
Chairman
r->
...................................
7Aif
Professor Rick L. Danheiser..
Thesis Supervisor
.
Professor Scott C. Virgil
....
2
-
..................
Y
ACKNOWLEDGMENTS
I would like to thank my friends, colleagues, and mentors for making my five years
of graduate school at MIT both a very enjoyable and unforgettable experience. First, I
would like to thank my research advisor, Rick L. Danheiser for his guidance and advice
throughout my graduate career. He is an outstanding teacher, and I believe both the
technical and research skills I have learned while working in his group will be invaluable to
me in my career as a chemist. (I also hold him fully responsible for my taste for gourmet
coffee which has developed over the past five years!) I must also thank Eric N. Jacobsen
for the opportunity to conduct research as an undergraduate in his laboratory at the
University of Illinois. It was his excitement about chemistry that first inspired me to attend
graduate school, and he has continued to provide support and excellent advice over the
years. Wei Zhang and Lisa Fields, both graduate students in his group at the time, also
helped to encourage me to pursue a career in science.
There are several former Danheiser group members that I wish to mention. Ray
Miller is a good friend and mentor whose advice I seek and value on many occassions.
And his wife, Ivy (not a former Danheiser group member--but it seems appropriate to
mention her at this point!) is a true friend and "shopping buddy" that I have really missed
since she and Ray moved to Connecticut!
John Kane, an excellent chemist, provided me
with many helpful ideas and I've always admired his ability to get a huge amount of work
done and yet not appear busy. John's words of advice: "the ball is your friend," have had
a major influence in my athletic activities at MIT.
I feel very lucky to have shared a bay with Annie Helgason for four years (so
lucky, note that she even gets her own paragraph!) We have shared many great times
together-both in and out of the lab-I can't possibly remember them all ('course that's
another story, but anyways...). She has been a great friend---always willing to listen to
whatever your problem may be---be it chemical or personal in nature, offering sound
advice, followed by, of course, accompanying you to the Muddy (or "yddum") for that
often needed Beer(s)! I admire her adventurous outlook on life and know that we will keep
in touch.
Over the years, Kathy Lee has been a faithful early morning aerobics/tennis partner
I'm really going to miss! She was a great travel companion on our trip to Washington, DC
and I hope we attend many more ACS meetings together in our future. Brenda Gacek is a
very good friend and is always willing to take a look at that complex NMR spectrum from
that unwanted side product from your reaction--for which I am grateful! As a fellow
midwesterner and former Jacobsen group member, we have a lot in common--who knows
someday we may be neighbors! I would like to thank Sandy Gould for all her help over
3
the years-from moving my furniture to proofreading my thesis. I wish her a fun summer
as captain of the toxic waste softball team (don't forget the beer!) and the best of luck both
finishing up here and as afuture member of the Jacobsen group.
My fellow fifth year, Alexandre Huboux-I know you must be disappointed that I
can't correctly pronounce your name...but not as sorry as I am that you don't like Lyle
Lovett (you're really missing something big!) I've enjoyed your friendship over the past
five years and your move into the computer room has expanded my musical tastes greatly
and made my final thesis-writing days more enjoyable! The best of luck in France!
Fari
Farooznia "cuuuuzzz", my other fellow fifth year, I wish you the best of luck at Ciba!
Mike Lawlor, I admire not only for his natural coaching ability and determination to
drink his own coffee, but also for one of the greatest lines I've ever heard: (in response to
RJN asking what Mike was getting him for Christmas)..."a muzzle". Which brings me to
Rob J. Niger, whose taste in music I actually really like...but does it have to be so loud???
I wish you the best of luck cleaving off that "#!*?**-ing" silicon group and remembering to
buy the coffee filters! Adam Renslo is an excellent addition to the Danheiser athletic teams
and will, I'm sure, be an outstanding IR-UV king. I thank Melanie Bartow and Matt
Martin for their patience and help with all the computer troubles I encountered during my
thesis writing endeavor. Matt, of course, I hope that your chemistry works amazingly well
but perhaps even more I hope that the RED SOX win the pennant at some point during
your graduate career! And to Dawn Bennett, I wish you the best of luck deciphering my
notebooks, results, etc... and most of all synthesizing many silylketenes!
I thank David Megan for the friendship and support he has provided even during
the most stressful, of times. I admire his optimistic outlook on life and enjoy his pop
quizzes of the elements of the periodic table! Heidi Erlacher and Jon Come, I would like to
thank you for making my time at MIT enjoyable (not to mention...that first year--SAFE! I
wouldn't have survived without you! ) Thanks to all the members of the Toxic Waste
Softball team---it was a lot of fun!
And a special thanks to all of you (you know who you
are---your eyes are probably still recovering) who proofread many-many pages of
chemistry for me--.-I am grateful!
4
To Mom and Dad
with love
5
I. TOTAL SYNTHESIS OF DAN SHEN DITERPENOID QUINONES
II. SYNTHESIS AND CHEMISTRY OF (TRIALKYLSILYL)VINYLKETENES
by
Jennifer L. Loebach
Submitted to the Department of Chemistry
on May 4, 1995 in partial fulfillment of the
requirements for the Degree of Doctor of Philosophy
ABSTRACT
Part I:
The preparation of gram quantitites of three naturally occurring and biologically
active quinones has been accomplished using an aromatic annulation strategy developed in
these laboratories based on the photochemical Wolff rearrangement. In only five steps,
(+)-Danshexinkun A was synthesized in optically active form via the annulation of a chiral
silyloxyacetylene with an ac-diazo ketone. Cyclization of (+)-danshexinkun A then
afforded (-)-dihydrotanshinone I, which could be easily dehydrogenated to tanshinone I.
Part II:
Silylketenes exhibit dramatically different properties from those found for most
ketenes. A widely applicable and productive route to a variety of substituted
(trialkylsilyl)vinylketenes has been developed based on the photochemical Wolff
rearrangement of (o-silyldiazo ketones. These vinylketenes have been found to be
remarkabley robust substances which are stable to silica gel purification and heating in
C 6 D6 at 80 °C for several days.
A systematic investigation of the reactivity of these vinylketenes has been conducted
to demonstrate their use as versatile four carbon building blocks. Substituted
(silyl)vinylketenes have been shown to undergo Diels-Alder reactions with activated
olefinic and acetylenic dienophiles. These vinylketenes also participate in a novel [4 + 1]
cycloaddition which provides a new route to 5-membered carbocyclic compounds.
Thesis Supervisor: Rick L. Danheiser
Title: Professor of Chemistry
6
TABLE OF CONTENTS
Part I
Total Synthesis of Dan Shen Diterpenoid Quinones Isolated from
Salvia Miltorrhizia
Chapter
9
1
Introduction and Background
10
Chapter 2
Total Synthesis of Danshexinkun A, Dihydrotanshinone I,
and Tanshinone I
22
Part II
Synthesis and Chemistry of (Trialkylsilyl)vinylketenes ("TAS-Vinylketenes")
Chapter
38
1
Introduction and Background: Vinylketenes
39
Chapter 2
Introduction and Background: (Trialkylsilyl)ketenes
52
("TAS-Ketenes")
Chapter 3
Synthetic Approaches to Substituted TAS-Vinylketenes
70
Chapter 4
Applications of TAS-Vinylketenes as Dienes in the Diels-Alder
96
Reaction
Chapter 5
TAS-Vinylketenes as Four-Carbon Components in New [4 + 1]
120
Annulation Reactions
7
Part III
Experimental Section
155
General Procedures
156
Materials
156
Chromatography
157
Instrumentation
158
Experimentals and Spectra
160
8
Part I
Total Synthesis of
Dan Shen DiterpenoidQuinones
Isolated from Salvia Miltorrhizia
9
CHAPTER
1
INTRODUCTION AND BACKGROUND
Introduction:
Dan Shen Diterpenoid Quinones
Dan Shen is regarded as one of the most important drugs in Chinese traditional
medicine. 1 Obtained from the dried root of the Chinese red-rooted sage Salvia miltiorrhiza,
today Dan Shen is used clinically for the treatment of heart disease, menstrual disorders,
miscarriage, hypertension, and viral hepatitis.lb Dan Shen also displays antipyretic,
antineoplastic, antimicrobial, and anti-inflammatory properties, and exhibits strong activity
against collagen-induced platelet aggregation.lb,2 Previously it has been shown that Dan
Shen's broad spectrum of biological activity is due to a number of interesting abietane
diterpenoid quinones. 2 - 3
To date, more than 50 related diterpenes have been isolated from Dan Shen. In
1934, Nakao and Fukushima first reported the isolation of three orange-red pigments from
Salvia miltiorrhiza and named these compounds tanshinone I, II, and III.4 (The WadesGiles system for Romanizing Chinese characters spells the drug "Tan-Shin"; the correct
pinyin pronunciation is "Dan-Shen", however the names for this class of compounds stem
from the older spelling, hence: tanshinones). Takiura found that tanshinone III was
actually a mixture of tanshinone II and cryptotanshinone (2). The structure of tanshinone I
(1) was determined by von Wessely
5
in 1940, and the following year Takiura elucidated
the structure of cryptotanshinone (2).6 Later, it was discovered by Takiura that in fact, the
1
2.
3.
4.
5.
6.
(a) Duke, J. A.; Ayensu, E. S. Medicinal Plants of China; Reference Publications, Inc.: Algonac, MI,
1985; Vol. 2, p 381. (b) Pharmacology and Applications of Chinese Materia Medica; Chang, J. M.; But, P.
P. H. Eds.; World Scientific Publishing Co.: Singapore, 1986; Vol. 1, pp 255-268.
For additional references documenting the biological activity of Dan Shen, see: (a) Footnotes 3-5 in Lee,
Choang, T. F.; Chow, H. F.; Chui, K. Y.; Hon, P. M.; Tan, F. W.; Yang, Y.; Zhong, Z. P.; Lee, C. M.; Sham,
H. L.; Chan, C. F.; Cui, Y. X.; Wong, H. N. C. J. Org. Chem. 1990, 55, 3537.
Thomson, R. H. Naturally Occurring Quinones; Chapman & Hall: London, 1987; Vol. 3, pp 624-629 and
references cited therein.
Nakoa, M.; Fukushima, T. J. Pharm. Soc. Japan 1934, 54, 154.
von Wessely, F.; Wang, S. Ber 1940, 79, 19.
Takiura, K. J. Pharm. Soc. Japan 1941, 61, 482.
10
originally isolated tanshinone II was composed of two compounds: tanshinone IIA (3) and
IIB (4).3 In 1962, elucidation of the structure of tanshinone IIA (3) was achieved by
Takiura 7 and in 1968, Thomson published the structure of tanshinone IIB (4),8 thereby
completing the assignment of structures for the major Dan Shen pigments.
I
0
0
Tanshinone
I
Tanshinone
I
n
v
11A
OH
Tanshinone
I
I
11B
I
I
Unfortunately, subsequent biological studies of these individual components of the
drug revealed that their activity did not reproduce all of the activity of the crude extract, so
efforts were then focused on the isolation of the less abundant components of Dan Shen.
Chinese chemists isolated the related compounds danshexinkun A (5), B (6), C (7), and D
(8), and (-)-dihydrotanshinone
I (9).9
HO
0
0
6
Danshexinkun B
Danshe, xinkun A
0
N
0
7
Danshexinkun C
.00 _-...
O
N
OH
uansnexInKun u
7.
8.
9.
HO
HO
OH
Os
(-J-U IIyuIUIidtl1IIIIllUII
.---
i
I
Takiura, K. T.; Koizumi, K. Chem. Pharm. Bull. 1962, 10, 112.
Baille, A. C.; Thomson, R. H. J. Chem. Soc. (C) 1968, 48.
(a) 5-7: Fang, C. N.; Chang, P.-L.; Hsu, T.-P. Acta Chem. Sinica 1976, 34, 197. (b) 8: Lou, H.-W.; Wu,
B.-J.; Wu, Mu, M.-Y.; Yong, Z. G.; Acta Pharm. Sinica 1985, 20, 542. (c) 9: Qian, M.-K.; Yng, B.-J.; Gu,
W.-H.; Chen, Z.-X.; Chen, X.-D.; Ye, X.-Q. Acta Chem. Sinica 1978, 36, 199.
11
Several of the purified Dan Shen components have been screened for biological
activity.
The tanshinones 1, 2, and 3 as well as danshexinkun A (5) and (-)-
dihydrotanshinone I (9) are effective coronary vasodilators.1 0 Antiplatelet aggregation
activity is reported for tanshinones 1, 2, 3, and (-)-dihydrotanshinone I (9).1
(-
Dihydrotanshinone I is highly active against the dermatophytic fungi Trichophyton rubrum,
T. mentagrophytes, T. tonsulans var. sulfureum, Microsporum gyseum, Sabourandites
canis, and Epidermophyton floccosum. 12 In the same report, cryptotanshinone (2), (-)dihydrotanshinone
(9), and tanshinone IIB (4) were all found to be active in vitro against
Staphylococcuc aureus and gram positive bacteria. Despite the bioactivity reported for
many of the pure individual compounds isolated from Salvia Miltiorrhiza, none approaches
the medical properties of the crude drug mixture. Unfortunately, further identification of
the most active components has been frustrated by the extreme scarcity of some of these
substances.
The three diterpenes shown below - danshexinkun A (5), (-)-dihydrotanshinone I
(9), and tanshinone I (1) - were selected as target structures for our synthetic investigation.
HO
N
0
0
OH
5
Danshexinkun A
0
0
0
9
(-)-Dihydrotanshinone I
1
Tanshinone I
The aim of this study was two-fold. One goal was to perfect the recently developed
"second generation" versionl 3 b of our annulation strategy and examine its application for
10. Chen, C.-C.; Chen, H.-T.; Chen, Y.-P.; Hsu, H.-Y.; Hsieh, T.-C. Taiwan Pharm. Assoc. 1986, 38, 226.
1 . Onitsuka, M.; Fujiu, M.; Shinma, N.; Maruyama, H. B. Chem. Pharm. Bull. 1983, 34, 1670.
12. Honda, G.; Koezuka, Y.; Tabata, M. Chem. Pharm. Bull. 1988, 36, 408.
13.
(a) Danheiser, R. L.; Nishida,
A.; Savariar, S.; Trova, M. P. Tetrahedron
Lett. 1988, 4917. (b) Danheiser,
L.; Brisbois, R. G.; Kowalczyk, J. J.; Miller, R. F. J. Am. Chem. Soc. 1990, 112, 3093.
12
R.
the synthesis of the tricyclic aromatic system found in compounds 5, 9, and 1. A second
goal was to develop efficient syntheses of these compounds practical enough to support the
preparation of gram quantities of each diterpene, thus facilitating the evaluation and
exploitation of their biological activity. Initial studies by David Casebier, a former member
of the Danheiser group, had previously led to a route to diterpenes 5, 9, and 1.14 My
research goals included the optimization of each synthesis, the complete characterization of
all synthetic intermediates and the final natural products, and the preparation of gram
quantities of the target diterpenes.
Previous Syntheses of Danshexinkun
A (5), (-)-Dihydrotanshinone
I (9),
and Tanshinone I (1)
Despite great interest in the isolation of natural products from Dan Shen, no total
syntheses of danshexinkun A (5) and relatively few total syntheses of tanshinone I (1) and
dihydrotanshinone
I (9) have been reported.
Prior to 1968, there were several
unsuccessful synthetic attempts on the tanshinones. 15 In 1968, the synthesis of the
incorrect isomer of tanshinone I was reported by King. 16 Later that same year, Baille and
Thomson reported the successful partial synthesis of several tanshinones.8 As outlined in
Scheme 1, tanshinone I (1) was produced from 8-methyl-3-phenanthrol (11) which was
obtained in 41% yield via the destructive distillation of podocarpic acid (10) from molten
sulfur. 1 7 The tricyclic intermediate 11 was oxidized to the hydroxy-para-quinone
12 and
the furan ring was then constructed via the introduction of an isopropenyl group at C-3
using -chloro-a-methyl propionyl peroxide followed by cyclization. Treatment of the
resulting ()-dihydrotanshinone I (9) with 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ)
effected dehydrogenation to afford quinone 1 which was identical with tanshinone I.
14.
15;.
16.
17.
Casebier, D. S. Ph.D. Dissertation, Massachusetts Institute of Technology, 1990.
Tanaka, S.; Kawai, S. J. Chem. Soc. Japan 1959, 80, 1183.
King, T. J.; Read, G. J. Chem. Soc. 1968, 5090.
Sherwood, I. R.; Short, W. F. J. Chem. Soc. 1938, 1006.
13
Scheme
1
OH
OH
OH
molten
sulfur
-
HO
2 C
12
3
59%
41%
H
N O.
1) HO3 SC6 H4N 2
2) FeC dil HCI, 0 2
3) 02, KOt-Bu, BuOH
1'
11
|l C0O
2
HOAc, reflux, 1 h
jt,
DDQ, PhH, reflux
0
44%
0
76%
1 Tanshinone
I
9 (±)-Dihydrotanshinone
I
Although this route begins with an expensive natural product as the starting
material, the target compounds are synthesized in only a few steps. Tanshinone I (1) is
produced in six steps from podocarpic acid 10 in 8% overall yield, and (+)dihydrotanshinone
I (9) is prepared in five steps beginning with 10 (11% overall yield).
In 1969, Kakisawa reported the first total synthesis of tanshinone I (1) utilizing a
Diels-Alder strategy to build-up the tanshinone framework (Scheme 2).18 The thermal
cycloaddition of 3-methylbenzofuran-4,7-quinone (13) (obtained in two steps from 2acetylresorcinol'9) and o-methylstyrene (14) afforded in 25% yield the Diels-Alder adduct
15. Kakisawa reported the formation of five colored products by tlc analysis in this
reaction, but was only able to successfully identify two of the products. The tetracyclic
quinone 15 was determined to be the major product, and the regioisomeric Diels-Alder
adduct was noted as a minor product. Hydrogenation of the furan ring in 15 using
platinum oxide in acetic acid afforded 16. Hydrolysis employing alcoholic potassium
18. Inouye, Y.; Kakisawa, H. Bull. Chem. Soc. Japan 1969, 42, 3318.
19. Walley, W. B. J. Chem. Soc. 1951, 3229.
14
hydroxide followed by heating with concentrated sulfuric acid in ethanol then provided (+)dihydrotanshinone
I (9) in 9% overall yield from quinone 13. (+)-Dihydrotanshinone
I
(9) was dehydrogenated with DDQ in 50% yield to produce tanshinone I (1) in 4% overall
yield. This Diels-Alder approach, although admirably convergent, suffers from very low
yields and the need to rearrange the para to the ortho quinone.
Scheme
2
O
100-110 C
+ I
\
O
0
10 h
I 2, PtO2, HOAc
o
25%
14
116
N
13
15
N
0
DDQ, PhH, reflux
50%
1
1. KOH, EtOH
2 conc. iS0 4
EtOH (1:1)
35% from 15
a
0
9
Tanshinone I
(+)-Dihydrotanshinone I
Recently, Snyder and coworkers have synthesized a number of tanshinones (but
not specifically dihydrotanshinone I (9) and tanshinone I (1)) using an ultrasound DielsAlder strategy.2 0 Their strategy is similar to that reported by Kakisawa but employs an
ortho-quinone as the dienophile. This modification eliminates extra steps late in the
synthesis and provides a concise, efficient route for the construction of the tanshinone
framework.
Imai has published a very different approach for the ring construction of tanshinone
I (1).21 His strategy assembles the ring skeleton via the sequence C-*D--A--B
3).
Commercially
available 2'-hydroxy-5'-methoxyacetophenone
(Scheme
(17) was first
brominated to provide a "handle" for further elaboration, and the furan ring was then
20
(a) Lee, J.; Snyder, J. K. J. Org. Chem. 1990, 55, 4995
Chem. 1992, 55, 5301.
21.. Imai, S. J. Sci. Hiroshima Univ. Ser. A 1971, 35, 171.
15
(b) Lee, J.; Li, J.-H.; Oya, S.; Snyder, J. K. J. Org.
formed using an intramolecular Perkin reaction. Alkylation of the Grignard reagent derived
from 19 with ethylene oxide gave a primary alcohol which was converted to 20 by
treatment with PBr3. The pentadienyl anion derived from resorcinol dimethyl ether by
Birch reduction was alkylated with 20 to yield the ACD intermediate 22. Acid catalyzed
cleavage of the methyl enol ethers and intramolecular Friedel-Crafts reaction generated the
B-ring. The ketone 23 was treated with methyl Grignard reagent and then dehydrogenated
using powdered sulfur. Cleavage of the methyl ether and oxidation with Fremy's salt
afforded tanshinone I (0.4 % overall yield).
Scheme
3
1) Br2, AIC13
2) BrCH2 CO2Et
K2C 0 3
acetone
MeO
OH
MeO
-
3) KOH, aq EtOH
4't7
I r
Rr
ROr
-
4Qa
Is
J
1) Mg°, Et2O
2) ethyleneoxide
3) PBr3, PhH
.
,
1) 2N H2 SO4
'
I
'
fio(
2) PPA, rt-*75
°C
OMe
21
38%
O
1)MeMgl, PhH
reflux,7h
2) S8, 210-215 °C
3h
4O
,
30%
MeO
0
1) H NNH2, KOH
DEG, 200 C,
then3% Na2 S 20 3
2) Fremy'ssalt
aqacetone
,
|
..
-;
O
--
NA
K2 HPO 4
_9%
C
O
a
A
I
B
1
Tanshinone I
Althought this is a lengthy synthesis (14 steps) for tanshinone I (1) in comparison
to the previous examples, this approach for the construction of the ring skeleton employs
some clever transformations which involve minimal deprotection and manipulation of
fimctional groups.
16
The most recent synthesis of tanshinone I (1) was reported by Huot and Brassard
(Scheme 4).22 Their route started with the phenanthrene catechol dimethyl ether 25, which
was synthesized from p-vanillin in nine steps in 16% overall yield. Cleavage of the methyl
ethers and oxidation to the ortho-quinone was accomplished in 60% yield to produce 26,
which was converted to the hydroxy para-quinone 11. Following the formation of the
silver salt of 11, treatment with crotyl bromide and heating produced 27 via a [3,3]
sigmatropic rearrangement of the crotyl ether. Zinc metal was employed to reduce the
quinone, and the three hydroxyl groups were protected as acetate esters. Oxidative
cleavage of the olefin generated the aldehyde and subsequent alcohol deprotection yielded
"quinonal" 28 which upon closure of the furan ring was transformed to tanshinone I (1).
Scheme
4
OMe
0
0
MeO
1)Ac 2O
1) BBr3, CH2 CI2
2) Ag20, MgSO4
g
2) NaOM
MeOH
70%
60%
25
26
OH
O
CHO
0
1) Zn, HOAc, Ac2O
N
2) Os0 4,dioxane
NalO4
3) NaOH
28
27
1) NaOH, DMSO
then HCI,SnCI2
2) dil HOAc
3) Jones Reagent
O
11
Tanshinone
22. Huot, R.; Brassard, P. Can. J. Chem. 1974, 52, 88.
17
This is the longest route to tanshinone I (1), employing 21 steps starting from pvanillin and resulting in 0.8% yield overall. This route requires two steps for the
conversion from the ortho- to para-quinone and several steps to incorporate the furan ring.
It is important to note that although we have chosen to present the published
syntheses of three Dan Shen natural products, considerable synthetic efforts have been
focused on several of the other interesting diterpenoid quinones that have been isolated
from Salvia miltiorrhiza. These efforts have utilized many of the strategies mentioned
above as well as other new innovative methods.23
Annulation Strategies for the Assembly of Aromatic Compounds
For the most efficient synthesis of highly substituted aromatic compounds, a
convergent and regiospecific strategy is required. Classical synthetic approaches to
substituted benzenoid compounds most often have employed linear substitution strategies.
Readily available aromatic derivatives are modified in a stepwise approach via electrophilic
and nucleophilic substitution reactions and more recently, directed-metal alkylation
reactions. 2 4 Disadvantages of some of these classical approaches include: (1) vigorous
WX.
StepwiseLinear~Sub~tepiution
pa
LinearSubstitution
- 0,
Strategies
L_
a-
y
,<
-
N~~
D1
n
Aromatic substitution
and directed
metalation reactions
-
X
R
R4
I
reaction conditions which may require additional protection/deprotection steps; (2) a lack
of convergent character, leading to lengthy routes; (3) problems of regiocontrol when
23. (a) Tateishi, M.; Kusumi, T.; Kakisawa, H. Tetrahedron, 1971, 27, 237. (b) Kakisawa, H.; Tateishi, M.;
Kusumi, T. Tetrahedron. Lett. 1968, 3783.
24. (a) Gawley, R. E.; Rein, K. In Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I.; Eds.; Pergamon
Press: Oxford, 1991, Vol. 1 pp. 459-476. (b) Snieckus, V. Chem. Rev. 1990, 879. (c) Gschwend, H. W.;
Rodriguez,
H. R. Org. React. 1979, 26, 1. (d) Townsend,
C. A.; Davus, S. G.; Christensen,
S. B.; Link, L.
C.; Lewis, C. P. J. Am. Chem. Soc. 1981, 22, 3923. (e) Narasimhan, N. S.; Mali, R. S. Top. Cur. Chem.
1987, 138, 63.
18
carrying out reactions on substituted aromatic systems. A more effective approach to
highly substituted aromatic compounds involves the application of annulation methods:
convergent strategies in which the aromatic system is assembled in a single step, with all or
most of the substituents in place. Aromatic annulation methods often enjoy significant
advantages over classical linear substitution strategies. Annulation methods avoid many of
the regiochemical ambiguities associated with classical substitution methods. Often, it is
possible to synthesize aromatic compounds with substitution patterns that would normally
not be accessible via the classical substitution approach.
r
Convergent
Annulation
nuStrategies
Strategies
L I
One-step
aromatic annulation
Acyclic
Precursors
~~~~~~~~~~~~~~~~~~~~R
1
]
Aromatic Annulation Strategies Based on Vinylketenes
A powerful aromatic annulation strategy has been developed in our laboratory that
is based on the addition of a vinylketene to an activated (heterosubstituted) or unactivated
acetylene. 13 The reaction involves the generation of the vinylketene via the 4 electrocyclic
ring opening of a cyclobutenone.
Recently, a "second-generation" version of this
annulation strategy has been developed which greatly expands the scope of the method.
In particular, this variant allows access to a variety of polycyclic
13b
aromatic and
heteroaromatic systems which were not previously available using the cyclobutenone-based
strategy. This new method is based on the generation of a vinylketene or an arylketene
using the photochemical Wolff rearrangement of an -diazo ketone. Scheme 5 outlines the
overall mechanistic course of the reaction. Irradiation of the a-diazo ketone induces Wolff
rearrangement to produce an aryl- or vinylketene which then combines in a [2 + 2]
19
c:ycloaddition with the acetylene. Further irradiation (or in some cases thermolysis) results
in 4T electrocyclic ring cleavage of the cyclobutenone ring to produce a dienylketene which
undergoes 6
electrocyclization
to afford a 2,4-cyclohexadienone,
which upon
tautomerization produces the desired aromatic product.
Scheme 5
I
1'N
%L
RI
R3
rX'
R4
y
R
3
RI
X
6 electron
electrocyclic closure
(and tautomerization)
photochemical
Wolff rearrangement
i
O'
C
R
2
R3
R4
R2
4 electron
electrocyclic
[2 + 2]
cycloaddition
x
cleavage
0
1R
R4
I
R
3
Synthesis of Diterpenoid Quinones Based on the Aromatic Annulation
Strategy
The pivotal step in our synthetic approach to the Dan Shen diterpenoids employs the
"second-generation" aromatic annulation strategy. The following retrosynthetic analysis
(Scheme 6) outlines our plan for the application of the annulation strategy to the assembly
of the key tricyclic intermediate 29, a precursor to (+)-danshexinkun A (5), (-)dihydrotanshinone I (9), and tanshinone I (1). Irradiation of the (a-diazo ketone 30
triggers a photochemical Wolff rearrangement to generate the arylketene 32, which upon [2
+ 2] cycloaddition with acetylene 31 would yield the cyclobutenone 33. Photochemical or
20
thermal cleavage of this cyclobutenone would lead to ketene 34, and electrocyclic ring
closure and tautomerization would then produce the desired tricyclic phenol 29 which
could be elaborated to afford danshexinkun A (5), dihydrotanshinone I (9), and tanshinone
I (1).
Scheme 6
-
N2
bPrSiO
HO
a
/
N.H ,| )
0
0
IN2
OH
CH 3
CH3
Danshexinkun
29
30
CH 3
A
photochemcal
Woff rearrangement
6 electron
electocydic dosure
(and tautomezation)
H
0
R3SI'
R3SiO
32
4 electron
R
CH3
CH 3
e/ecrcydic
[2+2]
cydoaddition
cleavage
CH 3
+
/Pr3 SiO-C-C J-
34
34
33
OSit-BuM2
31
Our approach is a particularly attractive route for the efficient assembly of this
tricyclic system and offers the potential flexibility to access a number of other Dan Shen
diterpenoid quinones.
14 ,2 5
25. Danheiser, R. L.; Casebier, D. S.; Huboux, A. H. J. Org. Chem. 1994, 59, 4844.
21
CHAPTER
2
TOTAL SYNTHESIS OF DANSHEXINKUN A,
DIHYDROTANSHINONE I, AND TANSHINONE I
Synthesis of the ax-Diazo Ketone Intermediate
The x-diazo ketone 30 required for the key aromatic annulation step was prepared
starting from 5-bromo-1-napthoic acid (35), a known compound which can be prepared on
a large scale by treating 1-napthoic acid with bromine in hot acetic acid according to the
procedure of Shoesmith and Rubli. 2 6 The bromo acid 35 was converted to the ketone 36
via the following one-pot procedure (eq 1). The bromo acid 35 was first deprotonated
with methyllithium to produce the carboxylate salt. Treatment of this salt with 2.1
equivalents of t-butyllithium effected halogen-metal exchange, and then 1.2 equivalents of
the alkylating agent methyl iodide were added followed by 2.0 equivalents of
methyllithium. After stirring for 20 h at room temperature, the reaction mixture was cooled
at 0 C and quenched with 20 equivalents of chlorotrimethylsilane. Upon workup with
dilute hydrochloric acid, the ketone 36 was produced in 83-86% yield after purification by
column chromatography.
C02 H
Br
Br
20 equiv Me 3SiCl
0 C
35
N2
CH 3U, THF, 0 C
t-BuLi
then CH31, -78 C
then CH3Li
-78 C - rt 20 h
83%
T
LIHMDS, TFEA°
then MsN,3
Et3N-H 20-CH 3 CN
0
(1)
84%
83%
36
CH3
CH 3
30
In the original study conducted by Casebier, this transformation was effected via
the treatment of a solution of lithium 5-methyl-l-napthoate with 3.0 equivalents of
26. Shoesmith, J. B.; Rubli, H. J. Chem. Soc. 1927, 3098.
22
methyllithium.
The reaction was quenched directly by slow addition to dilute hydrochloric
acid with rapid stirring. However, this procedure was found to be difficult to reproduce
without a significant amount (-30%) of the tertiary alcohol 37 being formed due to the
slow hydrolysis rate of methyllithium relative to the breakdown of the tetrahedral
intermediate into the methyl ketone and addition of methyllithium to the unmasked carbonyl
group.
HO
CH3
37
Tertiary alcohol formation has been reported as the most common side reaction
observed during the conversion of an acid to a ketone with a organolithium reagent. 2 7 The
amount of alcohol formed has been noted to be erratic and unpredictable. Methods devised
to minimize the formation of this side product include: quenching the reaction by addition
to dilute acid with rapid stirring and using several fresh portions of hydrolyzing medium;
employing exactly one equivalent of organolithium reagent with preformed lithium
carboxylates; using tetrahydrofuran as the solvent to increase carboxylate solubility; and
adding ethyl acetate or ethyl formate before hydrolysis to react with any remaining
organolithium reagent. After considerable study, we found that the carboxylate salt was
smoothly converted into the desired ketone by using the "Rubottom procedure"28 which
involves quenching the reaction mixture with a large excess of chlorotrimethylsilane.
Any
excess methyllithium reagent reacts quickly with the chlorotrimethylsilane and thereby
leaves no organometallic reagent to react with the ketone. This method provided the methyl
ketone 36 in 83-86% yield after purification by column chromatography.
Conversion of the ketone 36 to the ax-diazoderivative 30 was then achieved by
employing the improved "detrifluoroacetylative" diazo transfer method recently developed
27. Review: Jorgenson, M. J. Org. React. 1970, 18,1.
28. Rubottom, G. M.; Kim, C.-W. J. Org. Chem. 1983, 48, 1550.
23
in our laboratory. 2 9
The original Casebier procedure for this step used only 1.2
equivalents each of triethylamine and methanesulfonyl azide as well as a saturated sodium
bicarbonate wash in the workup. The reaction conditions were modified to conform to the
published method29 and 1.5 equivalents each of triethylamine and methanesulfonyl azide
were used. Also, it was found that the 10% sodium hydroxide wash stipulated in the
original procedure2 9 was indeed necessary to remove the sulfonamide byproduct which
was not separated from the diazo ketone when sodium bicarbonate was used.
The a-diazo ketone 30 exhibited strong characteristic IR absorptions at 2100 cm-1
(CN 2 stretching frequency) and 1615 cm-1 (carbonyl stretching frequency) and was
obtained as bright yellow needles, mp 84-85 °C, in 84% yield. This a-diazo ketone 30
was found to be stable to storage for extended periods of time (ca. 3 y at 0 °C).
Synthesis of the Siloxyacetylene
Annulation Component
Siloxyalkyne 31 was selected as the acetylene component for the pivotal aromatic
annulation step. Due to the ease of cleavage of a silyl ether group compared to an alkyl
ether group, siloxy derivatives are usually the preferred alkynes for our aromatic
annulationl
3
although both serve as effective ketenophiles.
Siloxyalkynes can be
conveniently prepared from carboxylic esters using the Kowalski reaction.3 0 a.b The
requisite optically active siloxyalkyne 313°Cwas prepared from commercially available (S)(+)-methyl 3-hydroxy-2-methylpropionate
(38) via (a) protection with t-butyldimethylsilyl
chloride, followed by (b) sequential treatment with lithiodibromomethane (generated in situ
from the treatment of dibromomethane with lithium tetramethylpiperidide), excess nbutyllithium, and triisopropylsilyl chloride according to the procedure of Kowalski
(Scheme 7).
29. (a) Danheiser, R. L.; Miller, R. F.; Brisbois, R. G.; Park, S. Z.; J. Org. Chem. 1990, 55, 1959. (b)
Brisbois, R. G., Ph. D. Thesis, Massachusetts Institute of Technology, 1990.
30. a) Kowalski, C. J.; Lal, G. S.; Haque, M. S. J. Am. Chem. Soc. 1986, 108, 7127. (b) Kowalski, C. J.; Lal, G.
S. J. Am. Chem. Soc. 1988, 110, 3693. (c) Evans, D. A.;; Sacks, C. E.; Kleschick, W. A.; Taber, J. R. J.
Am. Chem. Soc. 1979, 101, 6789.
24
Scheme 7
O
HO_0I_
HO
°C
imidazole
t-BuMe2SiCI
CH2C1I
- rt
O[.,)Me
O
.
...
.
12 h
100 %
38
'
_..
M
2.2 equiv UCHBr2
THF,-78°C;
4.9 equiv n-BuLi
-78 C - rt
39
I
20-35%
31
.1 equiv (
Casebier's yields for this reaction generally ranged from 47-53%. The isolated
yields that I obtained for pure siloxyacetylene were low (20-35%). This was believed to be
primarily due to the competitive formation of the isomeric silylketene identified by the IR
stretching frequency at -2100 cm-1 (C=O stretching). Chromatographic purification was
complicated by the presence of this silylketene, which has a very similar Rf to that of the
siloxyacetylene, and it was often necessary to run multiple (3-4) purification columns in
order to obtain uncontaminated siloxyacetylene 31. The siloxyacetylene was stable to
storage in a solution of dichloromethane at 15 C for several weeks; however after that
time, chromatography was necessary to purify the product before use.
With the
siloxyacetylene in hand, our efforts were now focused on the pivotal aromatic annulation
reaction.
Aromatic Annulation Step
The key annulation step was accomplished as shown in equation 2 by irradiating, in
a vycor tube, a degassed 0.35 M solution of the diazo ketone 30 and 1.4 equivalents of the
siloxyalkyne 31 in 1,2-dichloroethane with a low-pressure mercury lamp (254 nm). After
8 h, tlc analysis indicated that consumption of the diazo ketone 30 was complete. At this
point, two major products were evident by tlc analysis: the desired annulation product and
25
what was presumed to be the intermediate aryl cyclobutenone 33 (see Scheme 6, p. 13).
Polymer build-up on the inside of the vycor tube tends to impede the transmittance of light
and therefore the completion of the reaction. The reaction mixture was diluted with
additional dichloroethane and driven to completion by heating at reflux for 12 h.
Concentration and chromatographic purification furnished the tricyclic phenol 29 as a
yellow oil, [a]D 28.3
(c = 0.12, CHC13 ) in 70-75% yield. The 1H NMR spectrum of the
tricyclic phenol 29 exhibited a characteristic phenolic proton peak at 9.89 ppm as well as a
broad IR absorption at 3200 cm-1 (O-H stretching). The number of equivalents of
siloxyalkyne used for this annulation was decreased from the amount used in the original
Casebier study (2.0 to 1.4 equivalents) without affecting the yield of the reaction.
0°
i-Pr3SiO
OSit-BuMe2
hv (vycor)
CICH2CH2CI rt 8 h
then 90C 12h
Cad
30
OH
(2)
70-75 %
.OSi(I.Pr)
3
CH3
t-BUM 2SiO
29
31
Synthesis of Danshexinkun A
Cleavage of the silyl ether protective groups and oxidation to produce danshexinkun
A (5) was achieved in a single operation by exposure of 29 to the action of 2.2 equivalents
of tetra-n-butylammonium
fluoride in tetrahydrofuran (-78 °C to rt, 12 h) in the presence of
oxygen (eq 3). After a number of procedures were tried, the optimal conditions for
purification were found to be low temperature recrystallization from diethyl ether which
furnished (+)-danshexinkun A (5) in 41-58% yield as orange crystals, mp 200-201
200 oC),9 a [a]D +30
C (lit
(c = 0.01 CHC13 ). It is likely that a considerable amount of material
was lost during the purification. In earlier studies Casebier had obtained the diterpene in
91% yield following a similar protocol.
26
HO
)Sit-BuMe
2
O
2.0 equivn-Bu4NF
CHCI
2 2
then02
HOH
0
-78 °C-- rt
41-58%
(3)
5
danshexinkun A
Synthetic (+)-danshexinkun A (5) was characterized by 1H NMR,
IR spectroscopy.
1 3C
NMR and
The 1 H NMR spectrum of synthetic (+)-danshexinkun A (Figure 1) was
identical in all respects with the spectrum we obtained for an authentic sample of natural
(+)-danshexinkun A provided to us by Dr. Henry N. C. Wong (Table 1). We are grateful
to Professor Wong (Chinese University of Hong Kong) for providing us with authentic
samples of (+)-danshexinkun A, (-)-dihydrotanshinone I, and tanshinone I. The spectral
properties (1H NMR 300 MHz, CDC13 ) of synthetic (+)-danshexinkun A also closely
matched the published data (1 H NMR 60 MHz, CDC13). 9a
Recently, the enantiomer of (+)-danshexinkun A, (-)-danshexinkun A, has been
isolated from Salvia miltiorrhiza. 3 1 To date, no
13 C
danshexinkun A, therefore we have compared our
NMR data has been published for (+)-
13 C
NMR spectral data for synthetic (+)-
danshexinkun A to that reported for the enantiomer and found them to be in good
agreement (Table 2).
The melting point of our synthetic material (200-201 °C) was in excellent agreement
with the value we measured for natural (+)-danshexinkun A (198-200 °C) and with the
published value (200 °C).9 a A mixed sample of synthetic and natural (+)-danshexinkun A
was prepared by grinding equivalent amounts of each compound together. The mixed
melting point of this sample was observed at 197-200 °C. We had also planned to measure
a mixed melting point of a sample prepared from a solution of synthetic and natural
3:1. Ikeshiro, Y.; Hashimoto, I.; Iwamoto, Y.; Mase, I.; Tomita, Y. Phytochemistry, 1991, 30, 2791.
27
material. Unfortunately, due to the decompositition of the very small amount of authentic
natural (+)-danshexinkun we had obtained, we were not able to conduct this experiment.
To our knowledge, the optical rotation of (+)-danshexinkun A has not been
published in the literature.
The specific rotation measured for our synthetic (+)-
danshexinkun A was [a]D +30.00 (c = 0.01, CHC13). The magnitude of this value agrees
with the specific rotation measured (at a higher concentration) for (-)-danshexinkun A [a]D
-33.0 (c = 0.09, CHC13).3 1 An attempt was made to measure the optical rotation of the
authentic sample of (+)-danshexinkun A provided to us by Dr. Wong. The specific
rotation was found to be [a]D +12.50 (c = 0.2, CHC13). However, the purity of this
sample at this point at the time of the measurement is uncertain since when checked a few
days later it was found to have partially decomposed. Due to the small amount of sample
available, we were unable to repurify this material.
Although Fang and coworkers did not report the optical rotation of natural
danshexinkun A or its absolute stereochemistry,9 a we believe that it must be the S isomer.
Other researchers have referred to natural danshexinkun A as the S isomer,31 and this
absolute configuration is consistent with that found in a number of related diterpene
quinones isolated from the same plant. In particular, as discussed below, we have
converted our synthetic (+)-danshexinkun A into (-)-dihydrotanshinone I, the absolute
stereochemistry of which was established by comparison of its optical rotation to that of the
natural product of known absolute configuration.
28
;q
17
HO
0
OH
2
18
5
i
6~~~
11
0
10
14~
9
15
(.)-danshexinkunA
Table 1: 1 H NMR (300 MHz, CDC13) Spectral Data for Danshexinkun A
Rxvnthoti
VIIILIIV
-II- L (.1-nqnhoYinknin
I I YU·IY\·
U· AL-
_
Natural
hexinkiin -A
UHI (-Dan
I Y·'·---l---
1
2
3
8.05 (br, s)
8.05 (br, s)
4
5
9.43 (d, J = 9Hz)
6
7.64 (dd, J = 7, 9 Hz)
7
8
7.48 (d, J = 7 Hz)
9.42 (d, J = 9 Hz)
7.63 (dd, J = 7, 9 Hz)
7.46 (d, J= 7 Hz)
9
10
8.43 (d, J= 9 Hz)
8.27 (d, J= 9 Hz)
8.42 (d, J= 7 Hz)
8.26 (d, J= 9 Hz)
3.47 (m)
1.32 (d, J = 7Hz)
18a
2.74 (s)
3.51 (m)
1.34 (d, J = 7 Hz)
4.01 (dd, J= 8, 11 Hz)
18
3.87 (dd, J= 5, 11 Hz)
11
12
13
14
15
16
17
2.75 (s)
3.98 (dd, J= 8, 11 Hz)
3.89 (dd, J= 5, 11 Hz)
Table 2: 13C NMR Spectral Data for Danshexinkun A
_
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Synthetic (+)-Danshexinkun A
(75
MHz. CDCI')
---
Natural (-)-Danshexinkun A3 1
(100.4 MHz. CDC13-CD30D)
i
185.4
122.2
156.8
184.3
125.1
129.7
129.4
135.5
131.9
122.0
130.2
124.8
134.7
133.0
19.3
186.5
122.7
156.3
184.4
126.0
130.5
129.6
135.7
132.4
122.6
130.8
125.4
135.6
32.7
14.6
63.7
33.4
134.0
19.9
14.9
65.4
30
Synthesis of (-)-Dihydrotanshinone
I
As described by Fang et al., the treatment of (+)-danshexinkun A (5) with
concentrated sulfuric acid resulted in cyclization to afford (-)-dihydrotanshinone I (eq 4).9a
After column chromatography, (-)-dihydrotanshinone I (9) was isolated in 65% yield as
red needles, mp 221-222 °C (lit 224-225 OC).11
The specific rotation of synthetic (-)-dihydrotanshinone I was determined to be
[Ot]D-300
(c = 0.01, CHC13 ). This is in good agreement with the literature value reported
for the natural product [a]D -328 ° (c = 0.11, CHCl 3). 11 Due to the small amount of natural
dihydrotanshinone we were able to obtain, we did not measure its optical rotation.
The 1 H NMR spectrum of synthetic (-)-dihydrotanshinone I (Figure 2) was
compared to a spectrum we measured for the authentic sample and found to be in excellent
agreement (Figure 2 and Table 3). The
13 C
NMR spectral data for synthetic (-)-
dihydrotanshinone I (9) (75 MHz, CDC13) closely matched the spectral data published for
natural (-)-dihydrotanshinone I (100.4 MHz, CDC13-CD 3OD) 3 1 (Table 4). Resonances for
the ortho-quinone carbonyls were observed at 184.3 and 175.7 ppm.
The melting point of a natural (-)-dihydrotanshinone I provided to us by Dr. Wong
was 209-211
C, somewhat lower compared to the literature value (224-225 OC)ll and the
value measured for our synthetic sample (221-222 C). Examination of the 1 H NMR
spectra of the authentic sample reveals the presence of trace impurities which probably
account for the lower melting point.
HO
0
0
OH
O
CH3
0
H2 S0 4
EtOH
rt 30 min
65%
5
danshexinkun A
(4)
9
CH 3 (-)-dihydrotanshinone I
31
Figure 2: 1H NMR Spectrum of Synthetic (-)-Dihydrotanshinone I in CDC13
/
r . rl l · r
1H
, .......
r~~~~o
r~~~~~~o
9.0
a'.
--- -. .
.......
r.........rrlr ··r
I...................·r
Ilo~
.~
7,0
6.0
5.I
4.O
3.
2. 0
.0 PRO
NMR Spectrum of Authentic (-)-Dihydrotanshinone I in CDC13
r
J
I
.
r
i)
f
7.0
.
I.
.
f
4.
32
31.
1.
.
1
.
1,
O.
0
18
19
Table 3: 1H NMR (300 MHz, CDCl 3 ) Spectral Data for Dihydrotanshinone I
. ......
Svnthetic (-)-Tihvdrmntanqhinnne I
1
...... (-)-T)ihvlrntnnchinnnP,
I 1111
1T
Natiral
2
3
3.67 (ddq, J = 7, 10, 7 Hz)
49
4.44 (dd,J= 6, 10 Hz)
4.98 (app t, J = 10 Hz)
3.65 (ddq, J = 7, 10, 7 Hz)
4.42 (dd, J = 6, 10 Hz)
4.96 (app t, J = 10 Hz)
7.79 (d, J = 8 Hz)
8.33 (d, J = 9 Hz)
7.77 (d, J = 9 Hz)
8.33 (d,J = 8 Hz)
7.42 (d, J= 7 Hz)
4a
5
6
7
8
9
10
7.59 (dd, J = 7, 9 Hz)
11
9.32 (d, J= 9 Hz)
7.40 (d,J = 7 Hz)
7.58 (dd, J = 7, 9 Hz)
9.31 (d,J= 9 Hz)
1.42 (d, J = 7 Hz)
2.72 (s)
1.40 (d, J = 7 Hz)
2.70 (s)
12
13
14
15
16
17
18
19
Table 4:
13 C
NMR Spectral Data for Dihydrotanshinone I
Synthetic (-)-Dihydrotanshinone I
(75 MHz. CDCl)
1
2
184.3
175.7
Natural (-)-Dihydrotanshinone
- -184.3
175.7
34.7
81.6
34.8
81.7
7
8
9
10
120.3
132.0
134.9
128.8
130.4
11
125.0
12
13
14
15
16
17
18
19
134.8
132.1
120.3
131.8
135.0
128.9
130.4
125.1
134.8
132.2
128.3
126.2
118.4
170.4
18.8
19.8
3
4
5
6
128.2
126.1
118.4
170.6
18.8
19.9
33
I31
(100.4 MHz. CDCl-CDaOD'
I - -- -
Synthesis of Tanshinone I
Dehydrogenation of (-)-dihydrotanshinone I (9) was accomplished by treatment
with DDQ in benzene at room temperature to furnish tanshinone I (eq 5). Column
chromatography afforded tanshinone I (1) in 48% yield (74% yield based on recovered
starting material) as dark red needles, mp 226-227 °C (lit. 233-234 °C).l l The melting
point measured for an authentic sample of the natural product provided by Dr. Wong was
223-224 °C.
1H
NMR spectral data of both synthetic and authentic tanshinone I were
essentially identical (Figure 3 and Table 5) and in excellent agreement with the published
data. 11
13 C
NMR data for synthetic tanshinone I exhibited carbonyl resonances at 183.4
and 175.5 ppm. This spectral data was found to be consistent with published data for
similar ortho-quinone compounds.3 1
Due to a limited amount of synthetic material, this final transformation was tried
only once and afforded tanshinone I in 48% yield. However, Casebier was able to effect
this final step in 97% yield using similar DDQ conditions (1.9 equiv, benzene, rt, 36 h) and
this is considered to be the optimized yield for this step. 14
~~~0
N.
N
0~o
2.2 equiv DDO
benzene
0~rt95 h
~
48%••~~~~~~~~~
~48%
<~~
CH3
9
(-)-dihydrotanshinoneI
CH3
tanshinone I
34
(5)
I
8
4
1
I
Table 5:
1H
tanshinone I
19
NMR (300 MHz, CDC13 ) Spectral Data for Tanshinone
I
Synthetic Tanshinnne I
Nat1ral Tanchinnn T
7.22 (s)
7.28 (s)
6
7.78(d,J=9Hz)
7.81 (d,J = 8 Hz)
7
8
9
10
8.27 (d, J = 9 Hz)
8.30 (d, J = 9 Hz)
7.32 (d, J= 7 Hz)
7.52 (dd, J = 7, 9 Hz)
9.22 (d, J = 9 Hz)
7.34 (d, J = 7 Hz)
7.54 (dd, J = 7, 9 Hz)
9.24 (d, J = 9 Hz)
2.25 (s)
2.66 (s)
2.27 (s)
2.67 (s)
-1
2
3
4
5
11
12
13
35
Figure 3: 1H NMR Spectrum of Synthetic Tanshinone I in CDC13
I
§...
.,.V
1H
NMR Spectrum
..
.0
3:0
4:0
:.
'
of Authentic Tanshinone I in CDC13
rf
t . . . .'. . . .'.I .
I
I
J
3.
eI.a0.. P i
0
0.
.
fr
,
.
:)
.l.)..l3
7)
. . .3
l.l.:.3
O1'
l l.l~
...
5.,
,
36
I
.I
..
~
'.,
1....
1. I
·
......
0 ;q f
This is the first reported total synthesis of (+)-danshexinkun A (5). Only five steps
are required to produce (+)-danshexinkun A in 37-46% overall yield from 1-napthoic acid.
The synthesis of (-)-dihydrotanshinone I (9) establishes for the first time the absolute
stereochemistry of the furan ring. Our route to (-)-dihydrotanshinone I requires five steps
and affords the natural product under optimized conditions in 24-30% overall yield. From
(-)-dihydrotanshinone
I (9), only one additional step is needed to produce tanshinone I in
23-29% overall yield.
37
Part II
Synthesis and Chemistry
of
(Trialkylsilyl)vinylketenes
("TAS-Vinylketenes")
Part II of this thesis will focus on the synthesis
(trialkylsilyl)vinylketenes ("TAS-vinylketenes").
and chemistry
of
The development of a general and
efficient method for the preparation of these compounds based on the photochemical Wolff
rearrangement of a-silyldiazo ketones is presented as well as the results of an investigation
of their utility as four-carbon building blocks in new cycloaddition and annulation
strategies. The background relevant to our results will be presented in two parts: Chapter
1 - an introduction to the chemistry of vinylketenes with a discussion of their preparation,
reactivity, and use in synthesis, followed by Chapter 2 - an introduction to the chemistry of
silylketenes with a discussion of their preparation, stability, and use in synthesis.
38
CHAPTER
1
INTRODUCTION AND BACKGROUND:
VINYLKETENES
Introduction
Since the early part of this century, ketenes have been studied extensively and their
utility as versatile synthetic intermediates has been exploited for a variety of useful
transformations.
Pioneering studies on ketene chemistry were conducted in the laboratory
of Staudinger.32 These early investigations focused on the development of synthetic routes
to simple alkyl and arylketenes as well as the systematic examination of their chemistry,
including reactions such as dimerization, polymerization, and their addition reactions with
nucleophiles. Over the years, interest concerning the synthetic, mechanistic, and theoretical
aspects of ketene chemistry has increased greatly.3 3
Vinylketenes were suspected as intermediates in chemical reactions as far back as
1941.34 Relatively few vinylketenes have been isolated and characterized
35
due to their
natural tendency towards dimerization, polymerization, and oxidation. However, many
vinylketenes have been identified as transient intermediates in a number of reactions, and
also may be generated in situ for reaction with various substrates.
Generation of Vinylketenes
Presently, a number of methods exist for the preparation of vinylketenes. In most
cases the vinylketene is not isolated, but immediately reacts with a ketenophile present in
the reaction mixture.
32.
33.
34.
35.
(a) Staudinger, H. Chem. Ber. 1905, 38, 1735. (b) Staudinger, H.; Ott, E. Chem. Ber. 1908, 41, 2208.
(c) Staudinger, H.; Anthes, E.; Schneider, H. Chem. Ber. 1913, 46, 3539.
For recent reviews in the area of ketene chemistry, see: (a) "The Chemistry of Ketenes, Allenes, and
Related Compounds", Patai, S. Ed.; John Wiley and Sons: New York, 1980; and references cited therein.
(b) Hyatt, J. A.; Raynolds, P. W. Org. React. 1994, 45, 159. (c) Schaumann, E., Scheiblich, S. In
Methoden der organischen Chemie (Houben Weyl); Kropf, H., Schaumann, E., Eds.; George Thieme;
Stuttgart, 1993; Vol. E15/3, pp 2818-2880, 2933-2957.
Smith, L. I.; Hoehn, H. H. J. Am. Chem. Soc. 1941, 63, 1181.
(a) Wuest, J. D.; Madonik, A. M.; Gordon, D. C. J. Org. Chem. 1977, 42, 2111. (b) D6tz, K. H. Ang.
Chem. Int. Ed. Engl. 1979, 18, 954. (c) Datz, K. H.; Fugen-Kosterm, B. Chem. Ber. 1980, 113, 1449.
(d) Danheiser, R. L.; Sard, H. J. Org. Chem. 1980, 45, 4810.
39
A classical approach to the synthesis of ketenes involves the dehydrohalogenation
of an acid chloride. Payne applied this method for the first time to an a,3-unsaturated acid
chloride in 1966.36 In this early work, isopropenylketene was generated from 3-methyl-2butenoyl chloride with trimethylamine in the presence of ethyl vinyl ether. Both the ketene
dimer and [2 + 2] ketene-olefin cycloadduct were isolated. Since this initial study, a variety
of vinylketenes have been prepared employing the dehydrohalogenation strategy.37 For
example, Dreiding and coworkers have utilized the dehydrohalogenation method for the
synthesis of alkylvinylketenes and studied [2 + 2] vinylketene-olefin cycloadditions (eq
6).37d
EtN
0
Cl
^
r
150 C 4 h
(6)
sealed tube
[2 + 21
In 1977, Wuest reported the first isolation and characterization of a vinylketene. 3 5 a
Treatment of the acyl chloride 41 with triethylamine in benzene at 160 °C produced the
sterically shielded vinylketene 42 which was isolated by distillation at reduced pressure as
a yellow-orange liquid in 51% overall yield from 40 (eq 7).
o
II
C
CO2H
(CICO)
Et3N
160 0C 8.3
benzene
5-27°C
14h
82%
40
36.
37.
(7)L
benzene
62%
41
42
Payne, G. B. J. Org. Chem. 1966, 31, 718.
For several examples of the generation of vinylketenes via dehydrohalogenation, see: (a) Gelin, R.;
Gelin. S.; Dohnazon, R. Tetrahedron Lett. 1970, 3657. (b) Hickmott, P. W.; Miles, G. J.; Sheppard, G.;
Urbani, R.; Yoxall, C. T. J. Chem. Soc., Perkin Trans. 1 1973, 1514. (c) Rey, M.; Roberts, S.;
Dieffenbacher, A.; Dreiding, A. S. Helv. Chim. Acta 1970, 53, 417. (d) Rey, M.; Dunkelblum, E.;
Allain, R.; Dreiding, A. S. Helv. Chim. Acta 1970, 53, 2159. (e) Dondoni, A. Heterocycles 1980, 1547.
(f) Wuest, J. D. Tetrahedron 1980, 36, 2291. (g) Huston, R.; Rey, M.; Dreiding, A. S. Helv. Chim. Acta
1982, 65, 451. (h) Holder, R.; Dreiding, A. S. Helv. Chim. Acta 1983, 66, 2330. (i) Danheiser, R. L.;
Martinez-Davila, C.; Sard, H. Tetrahedron 1981, 37, 3943. (j) Danheiser, R. L.; Gee, S. K.; Sard, H. J.
Am. Chem. Soc. 1982, 104, 7670.
40
Unfortunately, the dehydrohalogenation method suffers in general from limitations
due to the triethylamine hydrochloride produced as a byproduct. This amine salt has been
noted to catalyze the polymerization of ketenes3 8 and isomerize the 3,y-doublebond of 2vinylcyclobutanones (products resulting from vinylketene-olefin [2 + 2] cycloadditions)
into conjugation with the carbonyl. 3 7h
An alternative method for generating vinylketenes involves the electrocyclic ring
opening of cyclobutenones
which can be effected under either thermal 3 9 or
photochemical4 0,4 5 conditions. In 1939, Smith and Hoehn observed the formation of xanaphthols when diphenylketene was heated with mono- and disubstituted acetylenes.41
Woodward and Arens later suggested the participation of cyclobutenones as intermediates
in this reaction (eq 8).42
r
-
0~~~~~~~
C
Ph
+
Ph
Ph
Ph
II
Ph
A
(8)
PPh
P
L
Ph
J
In the 1950's, Roberts et al. reported the thermal racemization of the optically active
cyclobutenone 43, which proceeds via a reversible electrocyclic ring opening (eq 9).43'
38.
39.
40.
4 1.
42.
44
For example, see (a) Brady, W. T.; Waters, O. H. J. Org. Chem. 1967,32, 3703. (b) Hansford, W. E.;
Sauer, J. C. Org. React. 1946, 3, 108.
For recent reviews, see: (a) Durst, T.; Brean, L. In Comprehensive Organic Synthesis; Trost, B. M., Ed.;
Pergamon Press: Oxford, 1991, Vol. 5, pp. 688-691. (b) Moore, H. W.; Decker, O. H. W. Chem. Rev.
1986, 86, 821. (c) Bellus, D.; Ernst, B. Angew. Chem., Intl. Ed. Engl. 1988, 797. (d) Marvell, E. N. In
Thermal Electrocyclic Reactions, Academic Press: New York, 1980, pp. 124-190.
For a discussion of the mechanistic differences which prevail under photochemical conditions, see:
Kikuchi, O. Bull. Chem. Soc. Jpn. 1982, 55, 1669.
(a) Smith, L. I.; Hoehn, H. H. J. Am. Chem. Soc. 1939, 61, 2619. (b) Smith, L. I.; Hoehn, H. H. J. Am.
Chem. Soc. 1941,63, 1175.
(a) Druey, J.; Jenny, E. F.; Schenker, K.; Woodward, R. B. Helv. Chim. Acta 1962, 45, 600. (b)
Nieuwenhius, J.; Arens, J. F. Recl. Trav. Chim. Pays-Bas 1958, 77, 1153. (c) Arens, J. F. Adv. Org.
Chem. 1960, 2, 191.
43.
44.
(a) Jenny, E. F.; Roberts, J. D. J. Am. Chem. Soc. 1956, 78, 2005. (b) Silversmith, E. F.; Kitahara, Y.;
Roberts, J. D. J. Am. Chem. Soc. 1958, 80, 4088. (c) Silversmith, E. F.; Kitahara, Y.; Caserio, M. C.;
Roberts, J. D. J. Am. Chem. Soc. 1958, 80, 5840.
For more examples of thermal electrocyclic ring opening to generate vinylketenes, see: refs. 13a, 37j, and
(a) Maahs, G. Angew. Chem., Int. Ed. Engl. 1963, 2, 690. (b) Maahs, G. Justus Liebigs Ann. Chem.
1965, 686, 55.
41
h
Ph
0
[H
1
c
H
EtOH
' I
Ph
43
OEt
Ch
H
(9)
Ph
Cl
Alternatively, the photochemical irradiation of cyclobutenones is known to promote
electrocyclic cleavage to generate vinylketenes at room temperature.
These mild
photochemical conditions have been employed for the generation of a variety of
vinylketenes. 4 5 For example, Mayr reported that cyclobutenone 44 opens under both
thermal conditions (refluxing cyclohexane, 81 °C) and photolytic conditions (irradiation in
carbon tetrachloride at 10 °C) to afford the vinylketene intermediate 45. This intermediate
was observed by its characteristic ketene band at 2097 cm-1 in the infrared spectrum and the
appearance of a yellow color which disappeared upon the addition of methanol to produce
the ester 46.46
Ph
0
Ph
hAor hv
Ph
.C
MeOH
Me
h
C
Me
Me
44
45
2 Me
46
Mayr has also reported the synthesis and trapping of methylprenylketene 48 via the
electrocyclic ring opening of 2,4,4-trimethylcyclobutenone (47). Cycloaddition of this
vinylketene with ethyl vinyl ether produced a mixture of stereoisomeric cyclobutanones (eq
11).45
45.
f
For more examples of photochemical electrocyclic ring opening to generate vinylketenes, see: (a) Barton,
D. H. R. Helv. Chim. Acta 1959, 42, 2604. (b) Baldwin, J. E.; McDaniel, M. C. J. Am. Chem. Soc.
1968, 90, 6118.. (c) Chapman, O. L.; Lassila, J. D. J. Am. Chem. Soc. 1968, 90, 2449. (d) Arnold, D.
R.; Hedaya, E.; Merritt, V. Y.; Karnischky,
L. A.; Kent, M. E. Tetrahedron
Lett. 1972, 3917.
(e) Huisgen,
R.; Mayr, H. J. Chem. Soc., Chem. Comm. 1976, 55. (f) Huisgen, R.; Mayr, H. J. Chem. Soc., Chem.
Comm. 1976, 57.
46.
Mayr, H. Angew. Chem., Intl. Ed. Engl. 1975, 14, 500.
42
C
1
]
OEt
EtO
i
47
0
+
EtO
48
The electrocyclic ring opening of cyclobutenones is often preferred over other
methods to generate vinylketenes because it does not produce a detrimental salt and allows
the ketene to be generated at low effective concentrations which helps minimize ketene
dimerization. In general, cyclobutenone ring opening serves as a mild and efficient method
for the preparation of vinylketenes.
As mentioned above, vinylketenes are often observed spectroscopically as reactive
intermediates. 4 7 In our laboratory, Kollol Pal conducted studies on the reversibility of the
photochemical generation of vinylketenes from cyclobutenones. Upon irradiation of a
solution of a particular cyclobutenone in perdeuterobenzene at temperatures ranging from 0
to 25
C, Pal was able to observe the presence
spectroscopy.
of a vinylketene
by
1H
NMR
4 7a
Vinylketenes have also been obtained via the photochemical Wolff rearrangement 4 8
of a,[3-unsaturated diazo ketones. As shown below in equation 12, irradiation of the diazo
ketone 49 initially afforded the vinylketene 50 which was then trapped in a [2 + 2]
cycloaddition with imine 52 to yield the -lactam 51.4 9 This is an attractive method for
vinylketene generation since the photochemical Wolff rearrangement can be induced at
room temperature and produces nitrogen gas as the only byproduct. Our laboratory has
also generated vinylketenes from a-diazo ketones via the photochemical Wolff
47.
48.
49.
(a) Pal, K., Ph. D. Thesis, Massachusetts Institute of Technology, Cambridge, MA 1987, pp. 69-77. (b)
Trahanovsky, W. S.; Suber, B. W.; Wilkes, M. C.; Preckel, M. M. J. Am. Chem. Soc. 1982, 104, 6779.
For reviews of the photochemical Wolff rearrangement, see: (a) Meier, H.; Zeller, K. P. Angew. Chem.,
Int. Ed. Engl. 1975, 14, 32. (b) Ando, W. In The Chemistry of the Diazonium and Diazo Groups; Patai,
S., Ed.; John Wiley and Sons: New York, 1978; Part 1, p. 458. (c) Regitz, M.; Maas, G. In Diazo
Compounds: Properties and Synthesis; Academic Press: Orlando, FL, 1986; p. 185. (d) Gill, G. B. In
Comprehensive Organic Synthesis; Trost, B. M., Ed.; Pergamon Press: Oxford, 1991, Vol. 3, p. 891.
Roedig, A.; Fahr, E.; Aman, H. Chem. Ber. 1964, 97,77.
43
rearrangement in our recently developed aromatic annulation strategy as discussed in Part I
of this thesis. 13b Irradiation of pyrazoles has also been shown to generate a-diazo ketones
which upon further irradiation afford vinylketenes as well. 50
Ph
Cl
,hAN=K
N2
Ci~CO2Me
ctC4M
Ph
-
hv
52
Cl
H
/
IC~
[2 + 2]
CI
49
(12)
O
Ph
51
In summary, three methods to produce vinylketenes have been described:
dehydrohalogenation of xa,3-unsaturated acid chlorides, electrocyclic ring opening of
cyclobutenones, and photochemical Wolff rearrangement of a,-unsaturated
diazo
ketones. There are also a number of vapor phase pyrolysis methods51 as well as other
photochemical methods (that do not involve the irradiation of cyclobutenones or diazo
compounds) by which specific vinylketenes have been generated. 5 2
Reactions of Vinylketenes
(1) Addition of Nucleophiles to Vinylketenes
In general, ketenes undergo addition reactions with nucleophiles such as water,
alcohols, amines, and carboxylic acids to yield acids, esters, amides, and anhydrides
respectively.
501.
5 1.
52.
Oftentimes there are simpler routes to these products and therefore
(a) Franck-Neumann, M.; Buchecker, C. Tetrahedron Lett. 1973, 2875. (b) Day, A. C.; McDonald, A. N.;
Anderson, B. F.; Bartczak, T. J.; Hodder, O. J. R. J. Chem. Soc., Chem. Comm. 1973, 247.
(a) Brown, R. F. C.; Butcher, M. Aust. J. Chem. 1969, 22, 1457. (b) Rousseau, G.; Bloch, R.; LePerchec,
P.; Conia, J. M. J. Chem. Soc., Chem. Comm. 1973, 795. (c) Schiess, P.; Radimerski, P. Helv. Chim.
Acta 1974, 57, 2583. (d) Schiess, P.; Heitzmann, M. Angew. Chem. Int. Ed. Engl. 1977, 16, 469. (e)
Terlouw, J. K.; Burgers, P. C.; Holmes, J. L. J. Am. Chem. Soc. 1979, 101, 225. (f) Holmes, J. L.;
Terlouw, J. K.; Vijfhuizen, P. C.; Campo, C. A. Org. Mass Spectrom. 1979, 14, 204. (g) Mohmand, S.;
Hirabayashi, T.; Bock, H. Chem. Ber. 1981, 114, 2609.
(a) Chapman, O. L.; McIntosh, C. L.; Pacansky, J. J. Am. Chem. Soc. 1873, 95, 244. (b) McIntosh, C.
L.; Chapman, O. L. J. Am. Chem. Soc. 1973, 95, 247. (c) Pong, R. G. S.; Shirk, J. S. J. Am. Chem. Soc.
1973, 95, 248. (d) Collins, P. M.; Hart, H. J. Chem. Soc. (C) 1967, 1197. (e) Griffiths, J.; Hart, H. J.
Am. Chem. Soc., 1968, 90, 5296. (f) Barton, D. H. R.; Quinkert, J. Chem. Soc., 1960, 1. (g) Quinkert,
G. Angew. Chem., Int. Ed. Eng. 1965, 4, 211. (h) Quinkert, K. R.; Schmieder, G.; Diirner, G.; Hache, K.;
Stegk, A.; Barton, D. H. R. Chem. Ber. 1977, 110, 3583.
44
nucleophilic additions to ketenes are not used a great deal synthetically. However, the
addition of nucleophiles is frequently used to confirm if a ketene was generated under the
reaction conditions.
In the case presented below (eq 13), the photochemical generation of
vinylketene 48 was established after the addition of dimethylamine to the reaction mixture
produced amide 54.50b
[
COMe
MeOC
N
hv
C
_
N
COMe
OO
53
[4+2](1
Diels-Alder
48
53
55
Me2 NH
0
NMe2
54
Similarily, Kollol Pal used this method in our laboratory to intercept vinylketenes
generated from the photochemical electrocyclic ring opening of cyclobutenones (Scheme
8). In these studies, tert-butylamine was employed as the nucleophile because it does not
contain a chromophore and therefore, would not interfere with the photochemical
transformation.
As shown below, the amine served as a very effective trap for the
vinylketenes. 4 7a
Scheme 8
0
NHt-Bu
tBuNH 2
96%
0
hv
t-BuNH2
100%
45
,NHt-Bu
NHt-Bu
(2) [2 + 2] Cycloadditions of Vinylketenes in Synthesis
The regiospecific [2 + 2] cycloaddition of ketenes with olefins is a common,
synthetically useful reaction which provides one of the most efficient routes for the
synthesis of four-membered rings. Mechanistically, the [2 + 2] cycloaddition is described
as a [2s + in2a] process, in which the ketene reacts in an antarafacial mode and the
ketenophile in a suprafacial mode. For cycloadditions of this type, ketenes are ideal
antarafacial components since one of their carbon atoms is an sp hybridized carbonyl
carbon which offers a minimum amount of steric hindrance. This results in a significant
decrease in the degree of crowding in the transition state favoring the [2 + 2] mode of
reaction.
A number of groups have utilized the [2 + 2] cycloaddition of vinylketenes in the
preparation of 2-vinylcyclobutanones.53 For example, Dreiding reported the generation of
methylvinylketene via in situ dehydrohalogenation and trapped it with a number of simple
olefins as illustrated below (eq 14).37d h
r
-1
Et3N
CHCI 3
20 ° C
0°
(14)
78%
L
J
.
1 65 35
(3) Annulation Strategies Based on Vinylketene [2 + 2] Cycloadditions
Over the past 15 years, several new annulation methods have been developed which
exploit the [2 + 2] reactivity of vinylketenes with both alkenes and acetylenes. Our
laboratory has been particularly interested in employing vinylketenes as four-carbon
components for the development of new synthetic approaches to a variety of carbocyclic
and heterocyclic systems (Scheme 9).
5:3
For examples, see:
Ref 13, 37d,h,i,j,
and 45f.
46
Scheme 9
ERZ0-i
a
1"__1
fy
[1,3]Sigmatropic
Rearrangement
RE
OH
OH
Electron-Rich
Alkenes
[
b
b
R
RE
1
1,3-Dienes
C
c
I
Acetylenes
R'-CC-CR
13-Dienes
3,3 Sraigmatropic
0
[
Rearrangement
~R~O
2
1:R2/
1'
IR
J
-
~~~~~~~~~~~OH
OH
R2
In 1981, our laboratory reported a new annulation approach to the synthesis of 3cyclohexenol derivatives via an alkoxy-accelerated rearrangement of vinylcyclobutanes
(pathway a). 37 i The vinylketenes required for this method were produced via
dehydrohalogenation of xa,-unsaturatedacid chlorides (using triethylamine in chloroform)
in the presence of excess ketenophile. The vinylketene generated reacts in a [2 + 2] fashion
with the olefin to produce a 2-vinylcyclobutanone, which upon addition of a suitable
nucleophile then undergoes an alkoxy-accelerated [1,3]-sigmatropic rearrangement to
produce the 3-cyclohexenol.
Overall, this process constitutes a powerful [4 + 2] annulation
which utilizes the four-carbon vinylketene unit for the synthesis of six-membered
carbocycles.
As outlined above (pathway b), the [4 + 4] annulation approach developed in our
laboratory for the synthesis of cyclooctadienone derivatives employs a 1,3-diene as one
four-carbon unit and a vinylketene as the second four-carbon unit.3 7 For this reaction, the
vinylketene is generated either by thermal electrocyclic cyclobutenone cleavage or via 1,4dehydrohalogenation of an xa,-unsaturated acid chloride. The vinylketene produced then
undergoes a regiospecific [2 + 2] cycloaddition with the conjugated diene to yield a 2,3-
47
divinylcyclobutanone
At elevated reaction temperatures, this
derivative.
divinylcyclobutanone rearranges to produce the eight-membered annulation product. Both
Dreiding3 7 g,5 4 and Tranhanovsky 4 7 b have reported this type of annulation strategy with a
vinylketene and cyclopentadiene which upon [2 + 2] cycloaddition rearranges to afford
bicyclo[4.2.1]nonadienones.
An annulation route to highly substituted aromatic systems based on the [2 + 2]
cycloaddition of vinylketenes and acetylenes (Scheme 9; pathway c) has been developed in
our laboratories. 13 This annulation method is described in detail in Part I of this thesis.
Related methodology for the synthesis of quinones has been developed independently in
the laboratories of Liebeskind55 and Moore56 using vinylketenes generated from modified
squaric acid derivatives
In 1975, D6tz5 7 reported a transition metal-mediated aromatic annulation that
employs Fischer carbenes for the synthesis of 1,4-hydroquinones. The mechanism of this
reaction (outlined in Scheme 10) involves the generation of a vinylketene intermediate
which is stabilized by complexation with chromium. Vinylketenes have been confirmed as
intermediates in this reaction through nucleophilic trapping experiments and the isolation of
stable compounds in several cases.5 7 b
54.
55.
56.
57.
Huston, R.; Rey, M.; Dreiding, A. S. Helv. Chim. Acta 1984, 67, 1506.
Liebeskind, L. S.; Iyer, S.; Jewell, C. F., Jr. J. Org. Chem. 1986, 51, 3065.
(a) Perri, S. T.; Foland, L. D.; Decker, O. H.; Moore, H. W. J. Org. Chem. 1986, 51, 3067. (b) Perri, S.
T.; Moore, H. W. Tetrahedron Lett. 1987, 28, 4507. (c) Moore, H. W.; Perri, S. T. J. Org. Chem. 1988,
53, 996.
For reviews, see: (a) D6tz, K. H.; Fischer, H.; Hofmann, P.; Kriessel, F. R.; Schubert, U.; Weiss, K.
Transition Metal Carbene Complexes; Verlag Chemie International: Deerfield Beach, FL, 1984. (b) D6tz,
K. H. Angew. Chem., Int. Ed. Engl. 1984, 23, 587. (c) Wulff, W. D. In Advances in Metal-Organic
Chemistry; Liebskind, L. S., Ed.; JAI Press Inc.: Grenwich, CT, 1989; Vol. 1. (d) Wulff, W. D. In
Comprehensive Organic Synthesis; Trost, B. M., Ed.; Pergamon Press: Oxford, 1991, Vol 5, pp. 10891106.
48
Scheme
10
91
R2
OCH 3
(CO)Cr
-co
I
W
R1
R2
r (CO)
F
4
phOCH3
,,,,,14
-
Cr(CO)
4
(4) Dimerization of Vinylketenes
The ease of preparation and isolation of ketenes is very dependent upon their
reactivity, particularly towards dimerization. Many ketenes are known to dimerize quite
readily. 3 3a .b For example, ketene can be handled as a pure colorless liquid at -78 C yet
dimerizes exothermically below room temperature to afford a [2 + 2] cycloadduct.3 3 b
Vinylketenes also dimerize, but produce [4 + 2] cycloaddition products as shown below.
Dreiding and coworkers described vinylketene dimerization as a new method to make xapyrones in the early 1970's (eq 15).37d
\-
r
CI
Et 3N
CHCi3
C
.
O
)
N
i
0
I
L
-J4
49
347/o
J
01
I o
(15)
(5) [4 + 2] Cycloadditions of Vinylketenes
Due to the normal tendency of vinylketenes to form [2 + 2] cycloadducts with
olefins, limited examples of the direct use of vinylketenes as [4 + 2] enophiles exist. In
1973, Day observed the reaction of the vinylketene 48 as the four-carbon diene in an
unexpected Diels-Alder reaction with the N=N bond of the diazo compound 53 (Scheme
11).50b
Scheme
11
MOC
:
COMe
r
N
r nMa
-
53
MeOC
C53
C
[4+2]1
Diels-Alder
L
53
48
I
46%
/
55
COMe
Similarily, diphenylketene 56 has been shown to react with the highly strained
imine, 2-phenylbenzazete (57) in a [4 + 2] fashion to afford the cycloadduct 58 (Scheme
12).58 This transformation is presumed to occur via a stepwise mechanism. Experiments
have demonstrated trapping of the zwitterionic intermediate 59 with water to produce the
amidoketone 60.
Control experiments have also been conducted to show that the
amidoketone 60 does not arise by hydration of the final product 58 or by the addition of
diphenylacetic acid to the azete 57.
58.
Rees, C. W.; Somanathan, R.; Storr, R. C.; Woolhouse, A. D. J. Chem. Soc., Chem. Comm. 1976, 125.
50
Scheme
12
_
_
Ph
Ph
56=O +
56
e-
1
I
I
of
I
J
.L
58
H2 0
Ph
OCHPh
COCHPh,
60
More recently, Dreiding has demonstrated the cycloaddition of alkyl- and aldovinylketenes to enamines to afford [4 + 2] cycloadducts (eq 16).59 Upon mild oxidation
the cyclohexenone 61 undergoes a Cope-type elimination to yield cyclohexadienone 62.
tC)
E
COCI,
m-CPBA
6)
L
Et3N, ItCH2C -
1_11~~r
-y67%
-
61
62
To date, the utility of vinylketenes as synthetic intermediates has been well
established and their reactivity in [2 + 2] cycloadditions has proven to be useful in the
synthesis of a variety of cyclic systems. However questions still remain such as: "Is it
possible to prepare stable, isolable vinylketene derivatives?" "Can the [2 + 2] reactivity of
vinylketenes be suppressed?" and "Can vinylketenes function in other types of synthetically
useful cycloaddition and annulation processes?" Answers to these questions will be the
focus of the remaining chapters of this thesis.
59.
Berge, J. M.; Rey, M.; Dreiding, A. S. Helv. Chim. Acta 1982, 65, 2230.
51
CHAPTER 2
INTRODUCTION AND BACKGROUND: (TRIALKYSILYL)KETENES
("TAS-KETENES")
The objective of the studies described in the remainder of this thesis was to
develop a general method for the synthesis of TAS-vinylketenes and to investigate their
utility as four-carbon components in new annulation strategies. This chapter reviews the
chemistry of simple (trialkylsilyl)ketenes and explains the basis for our expectation that
TAS-vinylketenes might be stable compounds that would function as useful four-carbon
building blocks in synthesis.
Properties of TAS-Ketenes
Silylketenes exhibit dramatically different properties from those found for most
ketenes. 6 0 Silyl substituents have the unique ability to stabilize ketenes and suppress
their natural tendency to dimerize and undergo [2 + 2] cycloadditions.
61
This
stabilization has been attributed to an electron-releasing effect of the silicon-carbon bond. Hyperconjugative donation of this -bond to the ketene 7r-system renders the
carbonyl less electrophilic and less reactive toward cycloadditions and nucleophilic
addition reactions (a).
Me 3 SI
60.
O.
a
+
(a)
Me3 ,
M,
C
b
(b)-
A
C,
Reviews: (a) Colvin, E. Silicon in Organic Synthesis Butterworths: London, 1982; Chapter 14. (b) Moreau, J.L. "Organometallic Derivatives of Allenes and Ketenes" In The Chemistry of Ketenes, Allenes, and Related
Compounds; Patai, S. Ed.; Wiley: New York, 1980, pp 401-413.
Inc. New York, 1995; 348-365.
61.
0
e 3 SI
(c) Tidwell, T. Ketenes John Wiley & Sons,
For theoretical calculations regarding the stability and reactivity of these compounds, see (a) Gong, L; Mc
Allister, M. A.; Tidwell, T. T. J. Am. Chem. Soc. 1991, 113, 6021. (b) Allen. A. D.; Tidwell, T. T. Tetrahedron
Lett. 1991, 32, 847.
52
Thus far, the use of silylketenes in synthesis has been relatively limited, and the
majority of studies have been conducted on the "parent derivative",
(trimethylsilyl)ketene. Our laboratory reported the first synthesis of a (silyl)vinylketene
and its participation in Diels-Alder reactions in 1980.35d In these reactions, the
vinylketene behaves as an electron-rich diene in which the carbonyl oxygen exerts a
strong directing effect thus controlling the regiochemical course of the cycloadditions (b).
Preparation of (Trimethylsilyl)ketene and Other TAS-Ketene Derivatives
(Trimethylsilyl)ketene
(63) is a colorless liquid that boils at 82 °C. It is unusually
stable for an aldoketene with respect to polymerization and decomposition. Samples of
(trimethylsilyl)ketene can be stored neat under nitrogen at room temperature for several
years without decomposition.
In 1965, (trimethylsilyl)ketene was first prepared via the silylation of
ethoxyacetylene to afford 64 followed by the thermally induced elimination of ethylene at
120 °C as illustrated below. 6 2
OCH
2 CH3
Me3 Si --.
12000C
______
120
M
3 SI--O
°C
Me
3 SI...
65%
CCMe=O
H.1[
64
65
C=
63
A recent improvement in this method has been reported which involves the
thermal decomposition of (tert-butoxytrimethylsilyl)ethyne (66) (eq 17).63 Elimination
62.
63.
Shchukovskaya, L. L.; Pal'chik, R. I.; Lazarev, A. N. Dokl. Akad. Nauk SSSR 1965, 164, 357.
Valenti, E.; Pericas, M. A.; Serratosa, F. J. Org. Chem. 1990, 55, 395.
53
of 2-methylpropene from 66 occurred slowly at temperatures as low as 50 C and
in 63% isolated yield.
C to produce (trimethylsilyl)ketene
instantaneously
at 100-110
Alternatively,
the silylketene was generated and trapped quantitatively
with a
nucleophilic amine (eq 17). The starting alkyne 66 can be obtained either via the
silylation of tert-butoxyethyne6 3 or via a 3-elimination/silylation process beginning with
(Z)- 1-bromo-tert-butoxyethene.
M. 3SI
Me___S__
--
64
OC(CH 3
3
Ph2 NH, CHCI3
50-55 °C, 11 h
50-55 C, 11I
O
main
(17)
67
66
The
NPh 2
Me3SI
advantage
butoxytrimethylsilyl)ethyne
of
thermal
the
elimination
of
(tert-
is that the silylketene can be generated at considerably lower
temperatures and in the presence of nucleophiles. The increased bulk of the t-butoxy
substituent helps prevent polymerization reactions and electrophilic attack which are side
reactions that commonly occur when the silylketene is generated from the less bulky
(trimethylsilyl)ethoxyacetylene
On the other hand, ethoxyacetylene
(64).
is
commercially available and more readily accessible than the t-butoxy derivative.
Anhydride pyrolysis is another method that has been applied for the preparation of 63,
although, this method reportedly affords (trimethylsilyl)ketene in low yield.6 5ab
Olah has prepared (trimethylsilyl)ketene via the dehydration of commercially
available (trimethylsilyl)acetic acid (68) (eq 18).66 Dehydration of this acid using
dicyclohexylcarbodiimide (DCC) and a catalytic amount of trimethylamine gave the
silylketene in 63% yield.
64.
65.
Pericas, M. A.; Serratosa, F.; Valenti, E. Tetrahedron 1987, 43, 2311.
(a) Kostyuk, A. S.; Boyadzhan, Zh. G.; Zaitseva, G. S.; Sergeev, V. N.; Savel'eva, N. I.; Baukov, Y. I.;
Lutsenko, I. F. Zh. Obshch. Khim. 1979, 49, 1543. (b) Kotstyuk, A. S.; Dudukina, O. V.; Burlachenko, G. S.;
Baukov, Y. I.; Lutsenko, I. F. Zh. Obshch. Khim. 1969, 39, 467. (c) Lutsenko, I. F.; Baukov, Y. I.; Kostyuk,
A. S.; Savelyeva, N. I.; Krysina, V. K. J. Organomet. Chem. 1969, 17, 241.
66.
Olah, G. A.; Wu, A.; Farooq, O. Synthesis 1989, 568.
54
DCC, cat Et3N
Me3 SKJIx00
Me3S
Et4h
_-SIO'
63%
,'·~OH
h
(18)
==o
(18)
63
68
in 33% yield as a
Barton has observed the formation of (trimethylsilyl)ketene
minor product formed during the pyrolysis of (trimethylsilyl)-4,5-dihydrofuran (69)
(Scheme 13).67 Formation of the silylketene is rationalized to proceed via homolytic CO bond cleavage of 69 to form the diradical species 70. Following the expulsion of
ethylene, (trimethylsilyloxy)acetylene (71) is produced and then undergoes salt-promoted
isomerization to (trimethylsilyl)ketene
(63).
Scheme 13
650o +-MC=C=O
---+
FVP
-
OSiMe3
H
69
69
SiMe3
OSIM.~
M
0
SiMe3
(5%)
(33%)
(49%)
C-O bond
cleavage
6
63
O SiMs
3 H
70
OSiMe
3
71
Kowalski has reported this same type of siloxyacetylene isomerization to produce
a silylketene. The ester 72 was homologated using the Kowalski method68 and the
ynolate anion generated was trapped with chlorotrimethylsilane
at -78 °C. The reaction
mixture was allowed to warm to room temperature to afford silylketene 75 as the only
product (eq 19). When the same ynolate was trapped at -78 °C and the reaction mixture
quenched with aqueous bicarbonate at -78 °C, a mixture of the siloxyacetylene 74 and
silylketene 75 (2:1 ratio) was produced. It appears that silylation kinetically occurs on
oxygen, but that isomerization to the more stable ketene occurs when the reaction is
67.
68.
Barton, T. J.; Groh, B. L. J. Am. Chem. Soc. 1985, 107, 7221.
Kowalski, C. J.; Haque, M. S.; Fields, K. W. J. Am. Chem. Soc. 1985, 107, 1429.
55
quenched at ambient temperature.6 9 This siloxyacetylene isomerization is limited to
small trialkylsilyl groups (i.e. Me3Si, Et3Si). Siloxyacetylenes with bulkier silyl groups
(such as i-Pr3Si) do not isomerize to afford the silylketene.
L
PhOEt
iTMP
0
CHBr;
Ph
72
OSiMe3
OS
Me 3SiCI
73J
I
(19)
-78 °C
rt
74
0
Pho~C
75
C,
SIMe3
While exploring the synthetic utility of (Z)-haloalkenes, Pirrung and Hwu
reported the quantitative conversion of alkene 76 to siloxyacetylene 77 (eq 20).70
However, upon close examination of Pirrung and Hwu's IR data, it was realized that the
compound isolated was not the siloxyacetylene 77, but was rather the silylketene 78
produced by isomerization of the initially formed siloxyacetylene.71
Br
OSiMe2t-Bu
76
2 equiv LDA
THF
0 C 30 min
TMSCI
100%
Me 3 Si--
OSiMe2t-Bu
]
77
0Sit
M9
3 Sie
SiMe2 Bu
78
69.
70..
Kowalski, C. J.; Sankar Lal, G.; Haque, M. S. J. Am. Chem. Soc. 1986,108,
Pirrung, M. C.; Hwu, J. R. Tetrahedron Lett. 1983, 24, 565.
71.
Danheiser, R. L.; Nishida, A.; Savariar, S.; Trova, M. P. Tetrahedron Lett. 1988, 29, 4917.
56
7127.
(20)
In 1979, Sakurai reported the first "satisfactory" method to make higher alkylsubstituted TAS-ketenes via the strategy shown below (eq 21).72 The alkoxyalkyne 79
was reacted with a slight excess of iodotrimethylsilane
(generated in situ from
hexamethyldisilane and iodine) to produce the alkyl(silyl)ketene 80 in moderate yield.
nrBu -
(CH 3) 3SilI
OCH2CH3
C-C =O
70'C 46h
57%
(21)
n-Bu
79
80
Alkyl- and arylsilylketenes have also been generated via the Wolff rearrangement
of corresponding
-silyldiazo ketones (eq 22).73 Wolff rearrangement has been induced
under photochemical conditions and using transition metal catalysis to afford silylketenes
in moderate to high isolated yields. This transformation has also been effected on a
variety of a-silyldiazo ketones with different trialkylsilyl groups.
hv ( > 280 nm)
benzene
O
rt
0
94%
C,,
Ph-~
(/Pr)3siC
(Pr) 3 Si /
CuOTf
benzene
rt
(22)
Ph
93%
82
Dehydrohalogenation, although a common method for ketene generation, does not
produce (trimethylsilyl)ketene in good yield.6
5b c
However, this method has been an
effective strategy for the synthesis of higher alkyl TAS-ketenes. For example, Tidwell
has generated (trimethylsilyl)ethylketene
(84) in this fashion in the course of his studies
on the stereospecific generation of silyl enol ethers from ketenes (eq 23).74
72.
73.
Sakurai, H.; Shirahata, A.; Sasaki, K.; Hosomi, A. Synthesis 1979, 740.
(a) Maas, G.; Bruckmann R. J. Org. Chem. 1985, 50, 2802. (b) BrUckmann, R.; Schneider, K.; Maas, G.
Tetrahedron, 1989, 45, 5517.
74.
Baigrie, L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc. 1985, 107, 5391.
57
Et3N
TH F
O
Me3Si
67 C
YlKC7I
Et
0
r Me3Si
2h
1
aBui,
Et=
83
Et2 O
-78 C 1 h
then Me3SiCI
84
Me 3 S,
OSiMe
Et
-78 -- rt 12h
78%
3
(23)
n-Bu
85
Zinc dehydrohalogenation has been employed by Brady for the synthesis of
87 (eq 24).75a Phenyl(trimethylsilyl)bromoacetyl
(trimethylsilyl)phenylketene
chloride
(86) was stirred with activated zinc at room temperature. After removal of the zinc salts,
the silylketene 87
was isolated by vacuum distillation in 41% yield.
Dehydrohalogenation using triethylamine was also attempted for the preparation of this
ketene but was unsuccessful due to carbon-silicon bond cleavage.
~
Ph~0
S~iACl
M
3h
Br
Zn-THF
41%
h
86
Ditz 7 5b,c
(24)
Me 3 SI,
rt
87
87
and Wulff75d have also isolated substituted TAS-vinylketenes in low to
moderate yields via the reaction of Fischer carbene complexes with silyl substituted
acetylenes.
Tidwell has prepared stable and persistent bis(silyl)-substituted bisketenes by
electrocyclic
ring opening of cyclobutenediones
bis(trimethylsilyl)acetylene
(eq 25).76
Cycloaddition of
(88) with dichloroketene (generated from trichloroacetyl
chloride using zinc-copper couple) gave dichlorocyclobutenone 89, which upon treatment
with concentrated sulfuric acid was hydrolyzed to afford cyclobutenedione 90 in 64%
yield over two steps. Ring opening of 90 could be induced either thermally (CDC13, 100
(hv, X=350 nm, 5
C, 48 min) to give
OC,
1 h, sealed tube) or photochemically
75.
(a) Brady, W. T.; Cheng, T. C. J. Organomet. Chem. 1977, 137, 287. (b) D6tz, K. H. Angew. Chem. Int. Ed.
Engl. 1979, 954. (c) D6tz, K. H.; Mihlemeier, J.; Trenkle, B. J. Organomet. Chem. 1985, 289, 257. (d) Tang,
P.-C.; Wulff, W. D. J. Am. Chem. Soc. 1984, 106, 1132.
(a) Zhao, D.-C.; Tidwell, T. T. J. Am. Chem. Soc. 1992, 114, 10980. (b) Zhao, D.-C.; Allen, D.; Tidwell, T. T.
J. Am. Chem. Soc. 1993, 115, 10097.
76.
58
bis(trimethylsilyl)bisketene (91) in excellent yield. This method has also been used to
prepare other bis(trialkylsilyl)-substituted bisketenes.
CCCOCI
Zn-Cu
Me 3 Si
Et20-DME
88%
SMe
88
0
Me3S i
Me3Si
CI
H 2 S0 4
55
12 min
73%
M3Si
MeSi
(25)
Me 3Si
Me3S
89
90
a (100%)
or
hv (86%)
Me3Sic
C
o cfC
SiMe3
91
Reactions of TAS-Ketenes
(1) Nucleophilic Additions
TAS-ketenes have been shown to readily undergo nucleophilic additions.6 b,77
Ruden has reported the use of (trimethylsilyl)ketene as a potent acylating agent for both
amines and alcohols. 77a (Trimethylsilyl)ketene reacted almost instantaneously with
hindered amines to form ax-silyl amides in quantitative yield (eq 26), whereas, the
reaction with alcohols such as tert-butanol was much slower (CC14 , 48 h, rt, 80% yield).
However, BF 3.OEt 2 has been found to strongly catalyze the addition of alcohols to
(trimethylsilyl)ketene.
The hindered tertiary alcohol 92, which could not be acylated
using standard reagents such as benzoyl chloride, acetyl chloride, and acetic anhydride
(even in the presence of DMAP), was successfully acetylated with (trimethylsilyl)ketene.
Desilylation was effected via the action of potassium fluoride in methanol to produce the
acetate 93 directly (eq 27).78
77.
(a) Ruden, R. A. J. Org. Chem. 1974, 39, 3607. (b) Lebedev, S. A.; Gervits, L. L.; Ponomarev, S. V.; Lutsenko,
I. F. J. Gen. Chem. USSR 1976, 46, 592. (c) Ponomarev, S. V.; Erman, M. B.; Lebedev, S. A.; Pechurina, S.
Ya.; Lutsenko, I. F. J. Gen. Chem. USSR 1971, 41, 122.
78.
(a) Danheiser, R. L. In Strategy and Tactics in Organic Synthesis, Lindberg, T., Ed.; Academic Press, Inc.:
Orlando, Florida, 1984; Vol. 1, Chapter 2. (b) Danheiser, R. L. Ph. D. Thesis, Harvard University, 1978.
59
i-Pr NH
I
cc 4 , rt
I
i-Pr2N
100%
Me 3SiK
"'1,/SiMe3
(26)
t-BuOH
0
BF3 OEt, CCI4
63
rt
2 min
SiMe 3
t-BuO
H
63, CCI 4
0 °C
then KF, MeOH
(27)
89%
92
93
Kita has found that a high degree of functionality in the substrate can be tolerated
when the addition of alcohols to (trimethylsilyl)ketene is catalyzed by zinc halides.
These functional groups include: carbonyls, acetals, thioacetals, epoxides, and olefins.79
In contrast, BF3 O-Et2 catalysis resulted in partial product desilylation with alcohols
containing carbonyl groups and also caused cleavage of acetal groups. The synthetic
utility of functionalized a-silylacetates has been demonstrated for the synthesis of
butenolides (eq 28). Similarly, Taylor has found that (trimethylsilyl)ketene is useful in
the synthesis of coumarins (eq 29).80
COH
0
63, ZnCI 2
CH2 CI 2
rt, .75 h
Ph
o1
)4
SiMe
3
(28)
91%
H3 C
CH3
CH3
0
OH
0
NaH
DMF
H
(29)
then 63, rt
98 %
NO2
79.
80.
0
NO2
Kita, Y.; Sekihachi, J. Hayashi, ; Y.; Da, Y.-Z.; Yamamoto, M.; Akai, S. J. Org. Chem. 1990, 55, 1108.
Taylor, R. T.; Cassell, R. A. Synthesis 1982, 672.
60
Tidwell has studied the addition of nucleophiles to (trialkylsilyl)ketenes and
bis(trialkylsilyl)bisketenes. 6 lb, 76 The reaction of bisketene 91 with water afforded the
mixture of anhydride products shown below (eq 30). Treatment of 91 with ethanol at 0
°C formed the monoketene 94, and an additional equivalent of ethanol reacts under
thermal conditions to produce the diester 95 (eq 31).
O
Me 3SiKC
H2 0
acetone
C
O
Me 3
(30)
q.
C
SIMe 3
0 1C
Me 3
91
(42%)
EtOH
91
Me 3SI
0 °C 15 min
O
(15%)
(8%)
,C
EtOH
C
A
Me 3 Si,
SiMe 3
(31)
I
C
;
-
Et2 OC
.CO 2Et
C
SiMe
Et2 OC-
3
95
94
a-Silyl ketones have been prepared by Kita in a one-pot procedure by the addition
of (trimethylsilyl)ketene to organocerium reagents followed by quenching with aqueous
ammonium chloride or alkyl halides (eq 32).81 Organolithium reagents are not suitable
for this reaction because proton abstraction is preferred over nucleophilic addition. In
fact, the addition of (trimethylsilyl)ketene
to a 0.01 M n-butyllithium solution at -100 °C
followed by quenching with chlorotrimethylsilane
silylketene 97 (eq
33).82
afforded in high yield the bis
Attempts to trap the ynolate with other electrophiles were
unsuccessful.
81.
82.
Kita, Y.; Matsuda, S.; Kitagaki, S. Tsuzuki, Y. Akai, S. Synlett. 1991, 401.
Woodbury, R. P.; Long, N. R.; Rathke, M. W. J. Org. Chem. 1978, 43, 376.
61
Me 3Sk,
H
0
PhCeC12
THF, -78
1.5 h
79%
HC=C=O
6
63
Me
3S
(32)
(32)
96
96
1) n-BuLi
Me3SIc
= C =O
2)
-100 C, THF
e100 CTF
3Si
80-90%
H'
C =O
(33)
Me3SI'
63
97
As an extension of the a-silyl ketone methodology, Kita has shown that
alkoxystannanes add to (trimethylsilyl)ketene, and the resulting a-(tributylstannyl)acetate
98 can be added to aldehydes and aldimines in the presence of TiC4 (eq 34).83 Notably,
the reaction with aldimines leads to the stereospecific preparation of syn-[-amino-a-silyl
esters which can be further elaborated to useful syn-amino diol derivatives.
PhCH=NBn
0
MeOSnBu
Me3SI.C==O CH 2 C2
H
H/
63
MeO
SnBU3
~~~~~~85%
98 SMe
TiC-78
-NBn
rt
-78 C -* rt
3
O
MeO
)
NBn
°C
99
Ph
SiMe 3
(2) Wittig and Related Reactions
(Trimethylsilyl)ketene has been used for the preparation of allenes. 77a, 84 For
example, olefination of silylketene 63 with a stabilized phosphorus ylide,
carboethoxymethylenetriphenylphosphorane, afforded the silyl-substituted allenic ester
100 in 85% yield (eq 35).77a This Wittig-type reaction occurred only with stabilized
phosphorus ylides, and unstabilized ylides were found to form complex mixtures. When
this reaction was conducted at room temperature, a mixture of the allene 100 and the
isomeric unconjugated (trimethylsilyl)acetylenic ester was formed. Acetylene formation
presumably occurs via isomerization promoted by the basic ylide.
83.
Akai, S.; Tsuzuki, Y.; Matsuda, S.; Kitagaki, S.; Kita, Y. J. Chem. Soc. Perkin Trans. 1 1992, 2813.
84.
Orlov, V. Y1; Lebedev, S. A.; Ponomarev, S. V.; Lutsenko, I. F. J. Gen. Chem. USSR 1975, 45, 696.
62
(34)
Ph 3P=CHCO 2 Et
Me3Si ,
HC--
H
-
Me 3 SiC
CH 2CI2 , -5 °C
8-
85%
63
H
C-C100
,H
C2Et
(35)
(3) [2 + 2] Cycloadditions
Silyl substitution greatly diminishes the natural tendency of ketenes to undergo [2
+ 2] cycloadditions.
As a result, cycloadditions of (trimethylsilyl)ketene with simple
olefins and dienes have been unsuccessful.77a,8 5 However, silylketenes do participate in
[2 + 2] cycloadditions with a few highly electron-rich olefins. Baukov first reported the
reaction of (trimethylsilyl)ketene with ketene diethyl acetal (101). Equivalent amounts of
the reactants were mixed together and heated. The cycloadduct 102 was formed in
moderate yield (eq 36).86 Brady has described similar [2 + 2] cycloadditions involving
(trimethylsilyl)ketene
and tetraalkoxyethylenes.
C= O
M3SIC,
90 "C 2h
+
(36)
(36)
+
-H =O.1
63
85
EtO 102
101
(Trimethylsilyl)ketene also undergoes [2 + 2] cycloadditions with aldehydes to
afford P-lactones.
87
Baukov 8 7a,b first reported the BF3yOEt2 catalyzed cycloaddition of
(trimethylsilyl)ketene with saturated aldehydes which results in mixtures of both cis and
85.
86.
87.
Brady, W. T.; Saidi, K. J. Org. Chem. 1980, 45, 729.
Zaitseva, G. S.; Baukov, Y. I.; Mal'tsev, V. V.; Lutsenko, I. F. Zh. Obshch. Khim. 1974, 44, 1415.
(a) Zaitseva, G. S.; Vinokurova, N. G. Baukov, Y. I. Zh. Obshch. Khim. 1975, 45, 1398. (b) Zaitseva, G. S.;
Vasil'eva, L. I.; Vinokurova, N. G.; Safronova, O. A.; Baukov, Y. I. Zh. Obshch. Khim. 1978, 48, 1363. (c)
Mead, K. T.; Yang, H-L. Tetrahedron Lett. 1989, 30, 6829. (d) Mead, K. T.; Samuel, B. Tetrahedron Lett.
1988, 29, 6573. (e) Brady, W. T.; Saidi, K. J. Org. Chem. 1979, 44, 733. (f) Maruoka, K.; Concepcion, A. B.;
Yamamoto, H. Synlett 1992, 31.
63
trans-2-oxetanones
(eq 37).
Ketones do not to undergo [2 + 2] cycloadditions
with
silylketene 63.
MeSi
63, BF 3 OEt 2
/-Pr04H
-,,
20 °C, 20 min
I-Pr
90%
H
0
3
H 90%
iPr
(37)
0
cis : trans
60: 40
Later, Brady reported cycloadditions of TMS-ketene and both saturated and ,Punsaturated aldehydes in the presence of BF3 -OEt2. The reaction of the a,-unsaturated
aldehydes with silylketene 63 afforded 2-oxetanones as indicated by IR; however, upon
distillation a [1,3]-shift of the organosilyl group occurred accompanied by ring opening to
yield trimethylsilyl dienoate esters.8 7e
Recently, a method for the stereoselective preparation of cis-substituted Plactones using the exceptionally bulky Lewis acid catalyst methylaluminum bis(4-bromo2,6-di-tert-butylphenoxide) (MABR) has been described by Yamamoto.8 7 f This method
is reported to work well for reactions of 63 with alkyl, aryl, and unsaturated aldehydes
(eq 38). When stoichiometric amounts of the catalyst were employed, desilylation
occurred followed by ring opening to afford (Z)-alkenoic acids.
0
Me 3S,. C=C=O
H/'EtH
+
cat MABR
-78 C, 2 h
CH2
82%
M
Me
(38)
2
Et
63
(4) Cyclopropane and Cyclobutane Formation
Baukov has shown that diazomethane adds to trimethylsilylketene to give two
types of products:
silyl-substituted cyclopropanones and cyclobutanones (Scheme
64
14).88a,b The treatment of (trimethylsilyl)ketene with an ethereal solution of an
equimolar amount of diazomethane at -130 °C produced (trimethylsilyl)cyclopropanone
(1.03) in 50% yield. The 3-membered ring product 103 was then reacted with a second
equivalent of diazomethane, and upon warming to -78 °C, ring expansion occurred to
yield a mixture of 2- and 3-(trimethylsilyl)cyclobutanones 104 and 105. Alternatively,
these isomeric 4-membered ring products were formed directly with two equivalents of
diazomethane at -78 C in 90% yield in a 40:60 ratio, with the 3-substituted product
favored. Treatment of the isomeric (trimethylsilyl)cyclobutanone mixture with methanol
makes it possible to obtain the pure 3-substituted isomer in 84% yield.89 This [2 + 1]
cycloaddition
with (trimethylsilyl)ketene
has also been effected
with
trimethylsilyldiazomethane, 88 d,e and the cycloaddition has been performed with other
as well.8 8f
(trimethylsilyl)arylketenes
Scheme 14
63
CH2 N2 , -130 C
50%
Me 3 SIC
103
2 ec
N2
78 °C
Et2 C
Me
3Si + 0
104
105
(5) [4 + 2] Cycloadditions
88.
(a) Zaitseva, G. S.; Bogdanova, G. S.; Baukov, Y. I.; Lutsenko, I. F.; J. Organomet. Chem. 1976,121, C1-C22.
(b) Zaitseva, G. S.; Bogdanova, G. S.; Baukov, Y. I.; Lutsenko, I. F. Zh. Obshch. Khim. 1978, 48, 131. (c)
Zaitseva, G. S.; Krylova, G. S.; Perelydina, O. P.; Baukov, Y. I.; Lutsenko, I. F. Zh. Obshch. Khim. 1981, 51,
2252. (d) Zaitseva, G. S.; Kisim, A. N.; Fedorenko, E. N.; Nosova, V. M.; Livantsova, L. I.; Baukov, Y. I. J.
Gen. Chem. 1987, 57, 1836. (e) Zaitseva, G. S.; Lutsenko, I. F.; Kisin, A. V.; Baukov, Y. I.; Lorberth, J. J.
Organomet. Chem. 1988, 345, 253. (f) Brady, W. T.; Cheng, T. C. J. Organomet. Chem. 1977, 137, 287.
89.
Tarakanova,
A. V.; Baranova, S. V.; Boganov, A. M.; Sefirov, N. S. Zh. Org. Khim. 1986, 22, 1095.
65
In 1993, Shioiri reported the first [4 + 2] cycloaddition reaction of a silylketene
with an electron rich diene. Silylketene 63 reacted smoothly with diene 106 in refluxing
benzene to afford the 2-pyranone 108 (eq 39).
Other aldoketenes with different
trialkylsilyl groups have been shown to be good heterodienophiles for this cycloaddition.
This transformation is a stepwise process with initial nucleophilic attack by diene 106 on
the sp-hybridized ketene carbonyl to produce the zwitterionic intermediate 107.
Hydrolysis of the reaction mixture at this point resulted in the isolation of ester 109 with
loss of the original ketene trialkylsilyl group. Prolonged reaction times or higher
temperatures promoted cyclization and upon hydrolysis the a,4-unsaturated lactone 108
was obtained. 9 0
o
1I
C
II
/C\
Me3 Si H
OSiMe3
benzene
80 C 3d
OSiMe3
51%
Me3 Si
0
(39)
CH3
63
CH3
106
SiMe3
108
IH2 O
OSOSiM
iM3
?
H3C
109
Preparation of (Trimethylsilyl)vinylketene
(Trimethylsilyl)vinylketene
dehydrohalogenation
90.
(112) was first prepared in our laboratories via a
reaction as outlined in the equation below. 3 5d
Ito, T.; Aoyama, T.; Shioiri, T. Tetrahedron Lett. 1993, 34, 6583.
66
DIBAL-MeLi
Et2 O-hex
CO2
KOH, MeOH;
1.1 equiv (COC) 2
pentane cat. DMF
0 C-
H3 C
SiMe3
rt 1.5 h
(40)
H,,
COCI
111
110
0.9 equiv Et3 N
pentane
A 15-24 h
0
Me3 Si
C
112
(39-50% overall yield)
Treatment of 1-(trimethylsilyl)propyne 11091 with diisobutylaluminum hydride
and methyllithium
(0 C, 0.5 h) in ether-hexane,
92
followed by reaction of the resulting
vinylalanate with anhydrous carbon dioxide yielded (Z)-2-(trimethylsilyl)-2-butenoic acid
in 68% yield. Exposure of the potassium salt of the acid to oxalyl chloride in pentane
containing a catalytic amount of dimethylformamide then produced the acid chloride 111
which was dehydrohalogenated without further purification. Thus, a solution of 111 in
pentane was added dropwise over 1-2 h to a solution of 0.9 equivalents of triethylamine
in pentane at 25 °C, and the resulting mixture was heated at reflux for 15-24 h and then
filtered with the aid of pentane.
Solvent was evaporated at -50 °C (0.5 mmHg), and the
residue was distilled at 25 °C (1 mmHg) and then again at 5 mmHg into a receiver cooled
at -78 C. In this manner (trimethylsilyl)vinylketene
(112) was obtained as a yellow-
green liquid in 39-50% overall yield (from 110).
(Trimethylsilyl)vinylketene
is a
relatively stable, isolable compound, and the purified vinylketene can be stored under
nitrogen in solution at 0 °C without appreciable decomposition for 1-2 weeks.
91.
92
(a) Eisch, J. J.; Damasevitz, G. A. J. Org. Chem. 1976, 41, 2214. (b) Uchida, K.; Utimoto, K.; Nozaki, H. J.
Org. Chem. 1976, 41, 2215.
Zweifel, G.; Steele, R. B. J. Am. Chem. Soc. 1967, 89, 2754.
67
[4 + 2] Cycloadditions of (Trimethylsilyl)vinylketene
(Trimethylsilyl)vinylketene
cycloadditions
(112) does not enter into typical [2 + 2]
with electron rich olefins.
Instead, it was found that
(trimethylsilyl)vinylketene participates in [4 +2] Diels-Alder reactions with a variety of
olefinic and acetylenic dienophiles. Ghosez later reported a similar [4 + 2] strategy
which employs N-silylated silylvinylketenimines which undergo cycloadditions with
olefinic and acetylenic dienophiles. 9 3
Our studies with (trimethylsilyl)vinylketene (112) illustrate the directing effect of
the carbonyl group in controlling the regiochemical course of the cycloadditions. For
example, reaction of silylvinylketene 112 with methyl propiolate produced methyl 3-
(trimethylsilyl)salicylate
(113) with the expected regiochemical orientation.
Protodesilylation of this adduct with trifluoroacetic acid afforded methyl salicylate (114)
in 78% yield (eq 41).
CO2CH3
112,tol
95°C 63h
45%
Si
OHMe
C
OCH
2
3
25MC, 24 h
(41)
78%
'
113
114
Diels-Alder addition of silylvinylketene 112 to olefinic dienophiles furnished
cyclohexenone derivatives as shown in equations 42 and 43. The addition of 112 to
naphthoquinone (115) afforded a mixture of several cycloadducts which could then be
oxidized to a single anthraquinone 116 (eq 44).
93.
Differding, E.; Vandevelde, O.; Roekens, B.; Van, T.; Ghosez, L. Tetrahedron Lett. 1987, 28, 397.
68
112, tol
95 °C 38 h
62%
0
Me 3 Si
(42)
112, CHCI3
Me 3SI
40 °C 24 h
N-Ph
74%
(43)
I-Ph
O
1) 112, CHCI 3
60 °C 41 h
0
2) 5 % KOH-EtOH
25 °C 1 h
28%
(44)
SIMe 3
0
O
116
115
OH
Unfortunately, at the time of the original investigation, no efficient and general
route to substituted TAS-vinylketenes was available. We have now developed a widely
applicable and productive route to substituted TAS-vinylketenes
based on the
photochemical Wolff rearrangement of a-silyldiazo ketones, as discussed in detail in
Chapter 3.
69
CHAPTER 3
SYNTHETIC APPROACHES TO
SUBSTITUTED TAS-VINYLKETENES
TAS-Vinylketenes via a Photochemical Wolff Rearrangement Strategy
One limitation of using the dehydrohalogenation method discussed in Chapter 2 for
the synthesis of (trimethylsilyl)vinylketene is the inability of this method to provide a
general route to substituted TAS-vinylketene derivatives. The application of this method to
the preparation of substituted TAS-vinylketenes is complicated by the regiochemical
ambiguity associated with the elimination step (eq 45).
1
,,°
1
0O
I
R3SI
2
..
I
..
C
-;(45)
l
o2
ri-
In general, the Wolff rearrangement of oa-diazoketones is an attractive option for
ketene generation since nitrogen is the only byproduct of the reaction, and the a-diazo
ketone substrates are usually very readily available. The aromatic annulation strategy
developed in our laboratoryl3 has established that the photochemical Wolff methodology
can be applied for the generation of vinylketenes, and Maas has also utilized this
rearrangement for the generation of (trialkylsilyl)aryl- and (trialkylsilyl)alkylketenes from
a-silyldiazo
ketones (see eq 22, p. 57). 7 3 Consequently, it was our expectation that the
photochemical Wolff rearrangement of a'-silyl-a'-diazo-a,[-unsaturated
ketones might
serve as a general and synthetically very useful method for the generation of TASvinylketenes.
At the beginning of the twentieth century, the Wolff rearrangement was discovered
accidentally by Ludwig Wolff. Wolff found that treatment of diazoacetophenone 117 with
70
water and silver oxide did not produce the expected ca-hydroxy ketone 118, but instead
afforded the rearrangement
product phenylacetic acid 119 (eq 46). 9 4 A few years later,
Schr6ter published his results of an analogous study. 9 5
0
y
0
Ao
N2
OH
.
118
A90
H20
117
e
f
(46)
C02H
119
Twenty years passed before the Wolff rearrangement became a synthetically useful
reaction. This was mainly due to a lack of convenient methods for the preparation of the xa-
diazo ketone starting materials. However, the discovery that a-diazo ketones could be
easily synthesized via the acylation of diazoalkanes with acid chlorides9 6 led to the
development of the Wolff rearrangement as a valuable reaction in preparative organic
synthesis. The reaction is commonly employed for Arndt-Eistert homologation of acid
chlorides, ring contraction strategies, as well as for ketene generation.
Wolff
rearrangement can be induced under thermal or metal catalyzed conditions as initially
described by Wolff. Alternatively, Horner discovered in 1951 that the Wolff reaction can
, This photochemical method of ketene generation is
also be initiated photochemically.97 48
often the procedure of choice since the rearrangement can be induced at room temperature.
94.
95.
96.
(a) Wolff, L. Justus Liebigs Ann. Chem. 1902, 325, 129; 1904, 333, 1; 1912, 394, 23.
Schr6ter, Chem. Ber. 1909, 42, 2346.
(a) Arndt, F.; Eistert, B.; Partale, W. Chem. Ber. 1927, 60, 1364. (b) Amdt, F.; Amend, J. Chem. Ber.
1928, 61, 1122.
97.
Homer, L.; Spietschka, E.; Gross, A. Liebigs. Ann. Chem. 1951, 573, 17.
71
Although the mechanism of the photochemical Wolff rearrangement has been
studied extensively, there is no widespread agreement on the mechanistic details connecting
reactants to products.
98
It is entirely possible that more than one mechanism is involved in
the Wolff rearrangement of different systems. Several possible mechanistic pathways are
illustrated in Scheme 15.
Scheme 15
R
N2
R2
120
hv in presenceof
triplet sensitizer
direct
21
1
191
-N2
/N2
N2
A
0
1
R'
R1
123
R2
1
0
0
R2
intersystem [
crossing
R2
124
R2
R'
125
126
H-atom
I
abstraction
c--!~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
RKHR 2
I
O-H, C-H
insertion
products
98.
128
(a) Murai, H.; Satarik, I.; Torres, M.; Strauz, O. P. J. Am. Chem. Soc. 1988, 110, 1025. (b) Mc Mahon,
R. J.; Chapman, O. L.; Hayes, R. A.; Hess, T. C.; Krimer, H.-P. J. Am. Chem. Soc. 1985, 107, 7597, and
references contained therein.
72
The pivotal question is whether the expulsion of nitrogen and substitutent migration
occur in a stepwise (via a ketocarbene) or concerted fashion under photochemical
conditions. A widely held view is that direct photolysis of an a-diazo ketone produces the
singlet excited state of the compound and brings about Wolff rearrangement (121 127).48C In fact, triplet-sensitized photolysis often suppresses Wolff rearrangement and
favors typical ketocarbene reactions such as H-abstraction and C-H and O-H insertion.
Migration of the R1 substitutent is known to occur in the singlet carbene species 121 but
not in the triplet carbene intermediate 122.99 Spin inversion may occur in both directions
when the singlet-triplet splitting is relatively small.
Evidence exists for the loss of nitrogen from the diazo ketone to form an
intermediate singlet ketocarbene (123).1°° Also, isotopic labeling experiments have been
carried out to prove that an isomerization can occur between ketocarbene 123 and oxirene
124 and carbene 125. For example, irradiation of the isotopically labeled diazo ketone
13 C
129 resulted in ca. 30% scrambling of the
label (eq 47).101 However, it should be
noted that the presence of an antiaromatic oxirene intermediate would only affect the
synthesis of a ketene if the aim was to prepare isotopically labeled material, since otherwise
both carbene 123 and 125 rearrange to the same ketene product.
O11
Ph
Ph
N2
Ph
hv (254 nm)
C
Ph
129
Ph
130a
Ph
Ph
(47)
130b
One well established fact is that the efficiency of the Wolff rearrangement depends
on the ground state conformation of the a-diazo ketone. Wolff rearrangement is favored
by the s-Z conformation of the diazo ketone (e.g. 120), in which the migrating group (R1 )
99.
100.
Tomiokla, H.; Okuno, H.; Izawa, Y. J. Org. Chem. 1980, 45, 5278.
(a) Meyer, H.; Zeller, K.-P. Angew. Chem. 1975, 87, 52. (b) Meyer, H.; Zeller, K.-P. Angew. Chem. Int.
Ed. Engl. 1975, 14, 32.
101.
(a) Fenwick, J.; Frater, G.; Ogi, K.; Strausz, O. P. J. J. Am. Chem. Soc. 1973, 95, 124. (b) de
Montellano, O. J. J. Am. Chem. Soc. 1980, 102, 7373.
73
is trans to the nitrogen leaving group that is then displaced by backside attack. The relative
efficiency at which several diazo ketones undergo rearrangement is illustrated in Table 6.48c
Table 6
% Yield of Photo-Wolff
Diazo Ketone
Major Conformer
Rearrangement
Products
0
Phv,
J
/N2
s-Z (96% at -40 °C)
97
s-Z (54% at -40 °C)
44
s-E
<3
s-Z
96
0
CH30-N2
f.Bfl~
Bu
t-Bu~
t.Bu
N2
In summary, the ability of the photochemical Wolff rearrangement to efficiently
provide ketenes under mild conditions from the requisite at-diazo ketones suggested that
this might be a very suitable method for synthesizing TAS-vinylketenes.
Our general route
for the synthesis of TAS-vinylketenes is illustrated below (Scheme 16). An attractive
feature of this approach is that a-diazo ketones can usually be obtained in one step from
methyl ketones in high yield via diazo transfer.29 Silylation using a modification of the
procedure of Maas73 followed by irradiation to induce Wolff rearrangement should then
afford TAS-vinylketenes.
74
Scheme
16
i-Pr2EtN
RIL
UHMDS,THF
CF3CO 2CH2CF3
RI
R2
then MsN3, E 3N
H2 0, CH3 CN
R2
N2
R3SiOTf
Et20-hexane
O
1
R
N
R2
N2
2
SiR3
hv
SIR3
RI
R2
Ketones
Synthesis of a-Diazo
All of the
-diazo
ketones
in this study were prepared
using the
"detrifluoroacetylative" diazo transfer strategy previously developed in our laboratory.29
This is a particularly valuable method for the convenient and efficient preparation of (a,3unsaturated a'-diazo ketones. An example of this transformation is shown below (eq 48).
0
LiHMDS
TFEA
THF -78 °C
0
-
132
O
CF3
0
MsN EI3N
H2 0-CH3cN
rt, 4 h
(48)
80%
131
133
132
Thus, reaction of the ketone substrate 131 with 1.1 equivalents of lithium
hexamethyldisilazide in THF at -78 C for 30 min produces the corresponding lithium
enolate, which is then acylated by exposure to 1.2 equivalents of trifluoroethyl
trifluoroacetate
(TFEA) at -78 °C for 5 to 10 min.
This trifluoroacetylating
agent is
commercially available and inexpensive, and generally leads to trifluoroacetylation at
carbon as desired. The resulting a-trifluoroacetyl ketone 132 is then treated at room
temperature for 4 h with 1.5 equivalents of methanesulfonyl azide in acetonitrile containing
1.0 equivalent of water and 1.5 equivalents of triethylamine. Column chromatography on
75
silica gel furnishes the desired a-diazo ketone in 80% yield. The preparation of 141 on a
10 g scale using this procedure has recently been described in Organic Syntheses.
10 2
The a-diazo ketones listed in Table 7 were prepared in very good yields using this
"detrifluoroacetylative" diazo transfer procedure. Reactions were generally done on a scale
that afforded between 500 mg and 2 g of product. The x-diazo ketones are stable, yellow
solids or oils that can be stored neat for long periods of time at 0 °C in the dark.
Table 7.
a-Diazo Ketones Synthesized via "Detrifluoroacetylative"
Diazo Transfer
Entry
a-DiazoKetone
% Yield
0
1
(134)
I
2
2
75-79
2,(134
N
N
(135)
80-97
".
3
phS4N
_
0
(136)
85-94
(137)
74-86
(138)
62-66
(139)
74
(140)
41'
(141)b
86
(133)
80
N2
5
j
"
O
6
Ph
,,
N
2
0
7
/S
N
0
8
Ph.N2
~9
.N2
Al~~
Yield notoptimized
aPreviouslyobtainedin 84% yieldbythis method(ref. 29a)
bPreviouslyobtainedin 87%yield bythis method(ref.29a)
cPreviouslyobtainedin 87%yield bythis method(ref.29a)
102.
Danheiser,
R. L.; Miller, R. F.; Brisbois,
R. G. Organic
76
Synthesis,
in press.
See also, Table 7, entry 8.
The spectra of a-diazo ketones include several characteristic features. 10 3 All of the
diazo compounds exhibit a strong IR absorption in the region from 1950-2300 cm-1 which
is assigned to either an N-N stretching mode or an asymmetric stretching of the C=N=N
cumulene system. The carbonyl group of a diazo ketone has decreased double bond
character due to the contributions of the resonance structures illustrated in equation 49.
This results in a relative shift of the carbonyl stretch (ca. 20-60 cm-l) toward lower
frequency, resulting in an absorption in the range of 1650-1700 cm- 1 .
0o
0
o
00(
N2
NNwR1J
2
R2
i
N2
R2
R2
120
142
143
The 1H NMR spectrum of a-unsubstituted diazo ketones exhibits a singlet
corresponding to the a-proton appearing in the 5.0-6.0 ppm range. This proton is
exchangeable in the presence of D2 0 and a catalytic amount of base. The
13
C NMR
spectrum is very important for characterizing a-diazo ketones. The resonance for the diazo
carbon appears at 50-70 ppm and the typical a,-unsaturated
diazo ketone carbonyl
resonance falls in the range of 180-190 ppm. The typical maxima for the ultraviolet
spectrum of a-diazo ketones is at 245-260 nm ( = 10,000-20,000), corresponding to the
7Tto t* transition of the diazo group.
All of the requisite methyl ketone starting materials are commercially available with
the exception of the precursors to 134, 137, 138, and 139. The preparation of 3-methyl3-penten-2-one (145) starting from commercially available tiglic acid is shown below (eq
50).
OH
2.0 equiv MeUi-LiBr
Et20
-78 °C - rt
90 min
0
(50)
85%
145
144
103.
For detailed discussions, see: (a) Ref. 48b pp 113-117. (b) Ref. 48c pp 3-64.
77
Conversion of the carboxylic acid 144 to the methyl ketone 14527 was initially
effected following a procedure by Frater 104 which involves the addition of 2 equivalents of
methyllithium to the acid at low temperature. The reaction mixture is then gradually
warmed to 0 °C, and then heated at reflux for 90 min. However, when we attempted to
repeat this procedure for the conversion of 144 to 145, the desired methyl ketone was
obtained contaminated with a significant amount (-10-15%) of the tertiary alcohol 146. In
this case, it was found that optimal conditions for the production of 145 did not involve
heating the reaction mixture to reflux. After warming to room temperature, the pure desired
methyl ketone could be isolated in 85% yield upon workup and purification by distillation.
The preparation of enone 151 is presented below (eq 51 and 52). The acetyl ylide
149 is prepared using the standard procedure. 105a Wittig reaction of ylide 149 with
hydrocinnamaldehyde
(150) provided 6-phenyl-3-hexen-2-one
Na 2C03
H20
620/
Ph3P, CHCI3
reflux, 1 h
14Cl
147'
148
0
H
Ph -
PPh3CI
P.PPh
3
(51)
149
149
0
149, THF
reflux 22 h
(52)
78% Phi
150
(151) in good yield.ll °b
151
The P,P-disubstituted methyl ketone 153 was prepared via a Horner-WadsworthEmmons reaction using commercially available dimethyl (2-oxopropyl) phosphonate as
described by Smith (eq 53).106 Methanol was used in this reaction (instead of ethanol as
recommended by Smith) and was found to readily add to the desired enone to produce 154
104.
105.
106.
Nussbaumer, C.; Frater, G. J. Org. Chem. 1987, 52, 2096.
(a) Ramires, F.; Dershowitz, S. J. Org. Chem. 1957, 22, 41. (b) House, H. O.; Respess, W. L.;
Whitesides, G. M. J. Org. Chem. 1966, 31, 35.
Smith, J. G. Ph. D. Dissertation, Harvard University, 1978, p. 130.
78
as the major product. However, the desired unsaturated ketone 153 was separated from
154 using column chromatography.
No attempts were made to optimize this reaction.
KOH, MeOH-H2 0
CH3COCH 2PO(OMe)2
0 C -- rt 16h
(53)
+
0
152
153
31%
154
45%
The three step synthesis of 5-benzyloxy-3-penten-2-one (158) is described below
(eq 54). Allylation of benzyl alcohol (155) was accomplished in excellent yield followed
by ozonolysis of the terminal double bond to afford aldehyde 157. The aldehyde was then
olefinated with acetyl ylide 149 to produce the methyl ketone 158 in 59% overall yield
from benzyl alcohol (155).
03
OH
Ph Ph155
155
CH2 =CHCH2 Br
DMF
NaH.
98%
Ph
98%
MeOH-ethyl acetate
Zn, AcOH
97%
156o
156
(54)
0
N.
Ph
NO
158
0
PPh3
0
149
THF
reflux, 7h
62%
Ph /O- sv
H
157
Synthesis of TAS-Diazo Ketones
Silylation of the a-diazo ketones was accomplished using a modification of the
method of Maas' 3 (eq 55). Maas has reported that the silylated a-diazo compounds
undergo protodesilylation very easily due to the trialkylammonium triflate (R3 NH+OTf- )
present in the reaction mixture (eq 56). We discovered that the yield of this silylation could
be improved by -25% by changing the solvent from Et20 to Et20-hexane (1:1) in order to
79
make the trialkylammonium
triflate less soluble and thus shift the equilibrium toward the
desired product. For example, the silylation of diazo ketone 159 afforded the desired
silyldiazo ketone 160 in 75% yield (eq 55), whereas using the Maas procedure the
silyldiazo ketone 160 was obtained in only 48-64% yield.
0
/Pr2EtN
iN2
Pr3 SiOTf
Et20:Hex
N2
I
r
75%
159
0
(55)
SI(.Pr)3
160
N
0
N4p~lP
Q. N
N
1
I
R3 NH+OTU-
H S(D
SIR
SIR
0N
(56)
R
x
As illustrated in Table 8, a number of TAS-diazo ketones were synthesized
including triisopropyl-, tert-butyldimethyl-, and triethylsilyl derivatives. All of the xasilyldiazo ketones prepared are new compounds that were previously unknown. Reactions
were typically conducted on a scale to produce between 100 mg to 1 g of product. aSilyldiazo ketones are unstable, yellow-brown oils that can be stored in solution for 2-3
days at 0 °C without significant decomposition.
In the 1H NMR spectrum of a-silyldiazo ketones, the singlet corresponding to the
a-proton of the diazo ketone precursor is absent and new resonances due to the trialkylsilyl
group are observed. The
13C
NMR spectrum exhibits a carbonyl resonance in the range of
185-196 ppm which is shifted downfield by -10 ppm when compared to the carbonyl
resonance of the ao-diazo ketone precursors. The IR spectrum of a-silyldiazo ketones
show strong CN 2 and C=O stretching absorptions in the 2040-2070 cm- 1 and the 1600-
1650 cm-1 regions, respectively. These absorptions are shifted to slightly lower frequency
as compared to the absorptions of the a-diazo ketones.
80
Table 8. Synthesis of TAS-Diazo Ketones
Entry
TAS-Diazo Ketone
R3 Si
Method of
Preparation
% Yield
0
1
N2
(161)
(162)
i-Pr3Si
Et3 Si
A
A
86-89
(163)
i-Pr3 Si
B
87
(164)
/-Pr3 Si
A
72
(165)
(166)
-Pr 3Si
Et3 Si
B
A
34-38
52
(167)
-Pr 3Si
A
93-100
iPr 3Si
A
0
(169)
i-Pr 3Si
A
16
(170)
i-Pr3Si
B
39
(171)
(160)
t-BuMe2Si
i-Pr3Si
A
A
53
75
70-84
SiR3
O
2
.
N2
SiR 3
4
4
Ph
0
N
N2
SiR3
O
N2
5
SiR 3
O
6
PhO.0
N2 (168)
N
SiR3
0
iRN 2
7
SiR 3
O
8
N2
PhN
SiR3
0
9
N2
SiR 3
Method A: Addition of base and the appropriate trialkylsilyltriflate to the diazo ketone in Et2 O-hexane (total
concn = 0.1 M) at 0 C)
Method B: addition of base and the diazo ketone in Et2 O-hexane to the appropriate trialkylsilyl triflate in
Et2 O-hexane (total concn = 0.02 M) at 0 °C)
The order of addition of reactants and the reaction concentration was found to be
crucial to the success of the silylation particularly for substrates with disubstituted double
81
bonds (entries 2, 3, 4, 6, 7, and 8). In these cases, initial experiments led to the formation
of the desired silyldiazo ketones contaminated with comparable amounts of a byproduct.
Upon further study, it was found that the side reaction leading to this byproduct
could be suppressed if (a) the silylation is conducted under more dilute reaction conditions
and (b) inverse addition of reactants is employed as described in Method B (addition of
base and the diazo ketone in Et2O-hexane to the appropriate trialkylsilyl triflate in Et 2 Ohexane (total concn = 0.02 M) at 0 °C). As indicated in Table 8, silyldiazo ketones with
disubstituted double bonds can be prepared in moderate yields using this Method B
procedure. It should be noted that for entries 6 and 7, only Method A was examined for
the synthesis of x-silyl diazos 168 and 169. It is believed that these silyldiazo ketones
would be obtained in higher yield if Method B was employed for their synthesis.
In most cases, the byproduct generated in reactions of disubstituted substrates (e.g.
entries 4,6, and 7') could not be separated by column chromatography. However in the
case of 1-diazo-4-phenyl-3-buten-2-one
(141),107 the products of the reaction were
separable and the desired silyldiazo ketone 170 was isolated in 40% yield (370 mg) along
with 240 mg of the byproduct 172 (eq 57). However, thus far we have been unable to
assign a definite structure to this compound. It is noteworthy that heating a purified sample
of silyldiazo ketone 165 at 40 °C for 22 h in C6 D6 led to the formation of a mixture of 165
and a comparable amount of the byproduct.
O
ph".,,,..,~
141
/
N
i-Pr2 EtN
-Pr3SiOTf
Et 0-hexane
02 °C- rt 25min
0
Ph
N2
172
10 Si(i-Pr3 )
170
40%
It is possible that this byproduct arises from a dipolar cycloaddition pathway. Maas
reports that oa-silyldiazo ketones can undergo a 1,3-(C--)O) silyl shift to produce
107,
The author gratefully acknowledges Raymond F. Miller for the preparation and characterization of the diazo
ketone 141.
82
diazoalkenes as reactive intermediates which have been trapped in the presence of
dipolarophilic alkenes such as N-phenylmaleimide (Scheme 17)108
Scheme
17
Si(-Pr)J
N-phenylmaleimide
Et20
40 °C 2 h
79%
(-Pr) 3SiQ
N,
N
Ph
The cycloaddition of diazo compounds to alkenes is a well known method for the
synthesis of pyrazoline rings. 10 9 It is possible that the diazoalkene 173 is generated and
undergoes cycloaddition with the double bond of an additional molecule of silyldiazo
ketone 170 to afford the indicated intermediate 1-pyrazoline. (eq 58). For this type of
cycloaddition, the indicated product is the expected regiochemical isomer where the
diazoalkane carbon is bonded to the 3-position of the a,13-unsaturated
carbonyl
component. 1 10
i-Pr EtN
O
Ph
2 ,
i-Pr3iOTf
Et20hexane
N_N2--eane
OSi(5Pr)38)
O
Ph2
~~~~141~1 Si(-Pr
3)
170
141
N-=N
Ph
(58)
173
170
172
a
L
108.
109.
110.
J
Munschauer, R. Maas, G. Angew. Chem. Int. Ed. Engl. 1991, 30, 306.
Regitz, M. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.; John Wiley & Sons: New York, 1984,
pp. 391-558.
Eberhard, P.; Huisgen, R. Tetrahedron Lett. 1971, 45, 4337.
83
The IR spectrum of the byproduct 172 (derived from silyldiazo ketone 170)
shows an N-H stretch at 3344 cm- 1 and also a CN 2 stretch at 2064 cm- 1 due to the
silyldiazo functionality in the molecule. Two carbonyl stretching frequencies are observed
at 1605 and 1630 cm-l1 . In the 1 H NMR spectrum, an amino hydrogen appears at 7.1 ppm
as a singlet and was found to be exchangeable with D2 0. The
13C
NMR spectrum shows
resonances at 195.0 and 49.5 ppm corresponding to an a-silyldiazo ketone functionality.
A second carbonyl resonance at 183.1, 11 sp2 -carbons between 121.9 and 156.3 ppm,
additional resonances at 74.6, 54.7, 31.6, and 22.6, and triisopropylsilyl carbons at 18.4
and 17.7 ppm were also observed in the
1 3C
NMR spectrum. The 1 H NMR spectrum of
172 exhibits a pair of doublets at 4.62 and 4.77 ppm with J = 3.6 Hz. The spectrum also
shows a doublet at 7.57 ppm with J = 15.8 Hz and a doublet at 6.60 ppm with J = 15.8
ppm. Each of the doublets described above integrates to one proton. One triisopropylsilyl
group is observed in the proton NMR spectrum, and additional multiplets appear between
7.2 and 7.7 ppm, integrating to ca. 15 protons.
To date, we have been unsuccessful in
assigning a structure to this compound.
Photochemical Wolff Rearrangement to TAS-Vinylketenes
Research directed toward determining the optimal conditions for the photo-Wolff
based generation of silylketenes was developed collaboratively with Dr. Kazu Takahashi.
As illustrated below, the rearrangment of ct-triisopropylsilyldiazoacetophenone
was initially
examined and optimal conditions for the transformation were established by Dr. Takahashi.
0
c
hv
PhJ
Si(-Pr)3
Si(i-Pr)
In general, the best results for the conversion of ca-silyldiazo ketones to silylketenes
were obtained by irradiating a 0.1 M solution of the a-silyldiazo ketone under a positive
pressure of nitrogen at 300 nm in a Rayonet RPR-100 photochemical reactor. The internal
temperature of the reactor chamber was maintained at 30-35 °C by use of a cooling fan.
84
Without the use of the fan, the chamber temperature rises to 65-75 C, resulting in
diminished yields of the desired products. The photochemical reactor was equipped with
sixteen low-pressure mercury vapor bulbs, which have a highly monochromatic output. It
was found that the highest yields and shortest reaction times were achieved when all sixteen
lamps were working. The rearrangement can also be induced at 0 °C when the reaction
mixture is irradiated in a Hanovia photochemical reactor, although yields were noted to be
slightly decreased using this method. Benzene, dichloroethane, hexane, and toluene all
proved to be suitable solvents for the reaction, whereas, acetonitrile was found to be
unacceptable.
The reaction also proceeds under the influence of 254 nm light, albeit in not as good
a yield. It is important to note that although all of the reactions subsequently described in
this thesis were conducted with 300 nm light in vycor tubes, these do not necessarily
represent the optimal conditions for every substrate. The UV spectrum of the a-silyldiazo
ketone of interest should be carefully evaluated prior to the choice of reaction conditions. A
strong absorbance at a wavelength significantly different than 300 nm should weigh heavily
in the determination of the exact experimental protocol.
The first ca,,3-unsaturated silyldiazo ketone studied was the cyclohexenyl derivative
160.
Optimal conditions for the generation of the TAS-vinylketene 174 involved
irradiating a degassed 0.1 M solution of the silyldiazo ketone in benzene in a vycor tube
using a low-pressure mercury lamp (300 nm) (eq 59). Concentration and chromatographic
purification furnished the desired ketene 174 as a stable, light yellow oil in 89% yield.
o
Sl(/-Pr
hv (300 nm)
N
(IP
3
b2
benzene(59)
89%
160
174
In general, the photochemical Wolff rearrangement produced the desired TASvinylketenes (100 mg to 1 g scale) in very good yield. In all cases, the reaction solution
85
was divided equally between two or three vycor tubes, which were simultaneously
irradiated in the photochemical reactor. Two sizes of reaction tubes were employed. The
standard vycor tube used in this investigation was 20-25 cm in length (9 mm O.D., 7 mm
I.D.). Larger tubes were generally 30-35 cm in length (12 mm O.D., 10 mm I.D.) and
were excellent for the preparation of larger amounts of material, but the greater volume of
the reaction solution often required longer periods of irradiation. Optimal results are
obtained when oxygen is scrupulously removed from the reaction tube. This can best be
accomplished by degassing the mixture with three freeze-pump-thaw cycles (-196 °C, <0.5
mm Hg). Alternatively,
when using larger tubes, we have found that it is acceptable to
degas the reaction tube by passing dry nitrogen gas through the solution for at least 10
minutes.
We have established that the photo-Wolff methodolology can be applied for the
efficient synthesis of TAS-vinylketenes with a wide range of vinyl and trialkylsilyl groups.
A variety of substituted TAS-vinylketenes have been prepared using this method as shown
in. Table 9. a-Silyldiazo ketones with a disubstituted double bond rearrange to produce
TAS-vinylketenes
(177, 179, 180, and 182) in only moderate yields. These are the
same substrates mentioned above that require Method B for silylation.
It is possible that
the yields for these photolysis reactions are low due to the instability of the silyldiazo
ketone starting material. These reaction mixtures also become dark red upon photolysis.
The dark color can prevent the transmittance of light to the silyldiazo ketones which could
then explain the poor conversion of silyldiazo ketone to ketene. The generation of the P,P-
disubstituted silylketene 181 from a-silyldiazo ketone 167 was unsuccessful. By tlc
analysis the reaction appeared to give one product, however, the yield of this product upon
isolation was very low (20-25%). The IR spectrum for this compound did show a strong
ketene stretch (-2100 cm-1), but the 1H NMR spectrum was not consistent with the desired
P,P3-disubstituted ketene. To date, the product of this reaction has not been identified.
86
Table 9. TAS-Vinylketenes via
Photochemical Wolff
Rearrangement
TAS-Vinylketene
EEntry
R3Si
Irradiation
Time (h)
%Yi.d
SiR3
1
C"I
(175)
(176)
-Pr3 Si
Et3 Si
4
3.5-7
79-80
65-73
(177)
i-Pr3 Si
2-3
54-61
(178)
/-Pr3 Si
(179)
(180)
i-Pr3Si
Et3Si
3
2.5
41
19-20
(181)
-Pr3 Si
2-3
0
(182)
-Pr3 Si
3
35
(174)
(183)
-Pr3 Si
t-BuMe2Si
4
4.5
89
58
SIR3
2
SiFi
2.5
81
SIR3
Ph
CO
0,
SiR3
7
ICo
As hoped, these TAS-vinylketenes have been found to be remarkably robust
substances and are even stable to silica gel purification.
The ketenes can be recovered in
-95% yield when the pure ketene is subjected to column chromatography. For example,
when silylketene 174 (0.055 g) was subjected to column chromatography on 10 g of silica
gel (elution with hexane) 95% of ketene 174 (0.052 g) was recovered unchanged.
87
1H
NMR was used as a tool to monitor the stability of TAS-vinylketenes
under
thermal conditions. The protocol developed for testing the ketene stability involved heating
a solution of the silylketene and 1,4-dimethoxybenzene(as internal standard) in deuterated
benzene under nitrogen in an NMR tube and monitoring by NMR the decomposition of the
ketene versus the internal standard. Decomposition of the (triethylsilyl)vinylketene 176
was observed
after
heating
in C6 D 6 at 80
C for 6-10 hours; however,
(triisopropylsilyl)vinylketenes 174, 175, and 177 were found to be considerably more
stable: no decomposition was seen after heating in C6 D6 at 80 °C for 3-4 days.
TAS-vinylketenes exhibit a number of interesting spectral characteristics (Table 10)
which make structure assignment a relatively straightforward task.l11 The IR spectra of
TAS-vinylsilylketenes show a strong diagnostic stretch near 2100 cm-1 due to the
symmetric stretching modes of the ketene backbone (C=C=O) and exhibit a principle
absorption band for the C=O stretch found near 2080 cm- 1. TAS-vinylketenes absorb in
the ultraviolet region with an absorption maxima near 220-240 nm. The 1H NMR
spectrum provides relatively little information since the TAS-vinylketenes do not contain a
diagnostic proton on the ketene double bond. However, the
contain some interesting features. 1 12
13 C
NMR spectrum does
The C-I carbon of the TAS-vinylketene is
extensively deshielded and produces a signal at low field near 180 ppm. On the other
hand, the C-2 carbon gives a very high field signal near 20 ppm, due in part to the
contribution of the resonance structure 184a illustrated below.
Me3Si
MeSi
2
184
111.
112.
c
184a
For spectral characteristics of ketenes see Ref. 33a pp 169-188.
Grishin, Y. K.; Ponomarev, S. V.; Lebedev, S. A. Zh. Org. Khim. 1974, 10, 402.
88
Table 10.
Spectral Characteristics of
(175)
3-Methyl-3-penten-triisopropylsilylketene
IR (film):
2941, 2881, 2081, 1461, 1381, 1291, 1181,
1071, and 1021 cm - l
IV max (hexane):
229 nm (e = 9000)
13C
NMR (75 MHz, CDC13 ):
18.6
12.5
O1
22.4
9 H
CH
119.5
18.8
.s
CH3
14.0
1H
I
NMR (300 MHz, CDC13 ):
-
l
1.10(d, J U.4 Hz)
1.16-1.27(m) s
0ac
g
H .l (q,J=6.7 Hz)
CH3
CH 3 1.u (d J
1.16(3)
6.7 Hz)
Il .
A,
I I I I I I I I
-
6--]
- F -f-
6
-
I
t
I4
5
I
I
I
I
I
I
I
3
4
89
J
II
I'
1
,
,
2
tA.I.
A
, I,
,1
,
- 1 I
1
r
I
Alternative Routes to TAS-Vinylketenes:
Preliminary Studies
(1) TAS-Vinylketenes via Electrocyclic Ring Opening of Trialkylsilylcyclobutenones
One route to the parent (trimethylsilyl)vinylketene
112 (see eq 40, p. 67 ) examined
in the original studies on this compound involved the electrocyclic ring opening of
silylcyclobutenone 185 which was hoped to be available via a [2 + 2] cycloaddition of
dichloroketene (189) and trimethylsilylacetylene (eq 60). Dichloroketene was chosen for
this cycloaddition because ketene itself is unreactive toward most alkenes and alkynes in [2
+ 2] cycloadditions. 113
Unfortunately, the cycloaddition of 188 and 189 afforded a
mixture of the desired (trialkysilyl)cyclobutenone 186 and (predominately) the undesired
P-silyl derivative 187.119
Me3S3Si
Me3
Me3 S
°
Me3Si
CIC
184
185
186
Me3Si
188
189
C
CI
187
[2 + 2] Cycloadditions of dichloroketene with alkynylsilanes has been shown to
proceed with high selectivity to produce the a-silyl cyclobutenones when the alkynylsilanes
are substituted with directing groups such as phenyl and ethoxy."
4
In principle, we
believed it would be possible to access substituted TAS-vinylketenes using the electrocyclic
ring opening strategy by employing suitably substituted acetylenes such as 191 (eq 61).
Silylation of ethoxyacetylene
190 proceeded in excellent yield to afford
trimethylsilylethoxyacetylene 77 a which was then added to dichloroketene (generated via
113.
114.
(a) Dehmlow, E. V.; Chem. Ber. 1967, 100, 3829. (b) Knoche, H. Justus Liebigs Ann. Chem. 1969,
722, 232. (c) Krebs, A.; Kimling, Justus Liebigs Ann. Chem. 1974, 2074. (d) Morita, N.; Asao, T.;
Kithara, Y. Chem. Lett. 1972, 927. (e) Wong, H. N. C.; Sondheimer, F.; Goodin, R.; Breslow, R.
Tetrahedron Lett. 1976, 2715. This method has also been successfully applied to cycloadditions
involving simple alkynes: Hassner, A.; Dillon, J. Synthesis, 1979, 689.
Danheiser, R. L.; Sard, H. Tetrahedron Lett. 1983 24, 23.
90
dehydrohalogenation of dichloroacetyl chloride) to produce the cycloadduct 192 in 43%
yield (lit. 86%).114 The next step in our scheme required the reductive dechlorination of a
4,4-dichlorocyclobutenone, which is considerably more difficult than in the case of the
corresponding saturated derivatives, though it can be accomplished using procedures
developed independently in our laboratory11 5 and that of Dreiding.1 16 Unfortunately, we
were not able to achieve clean dechlorination of the 2-silyl-dichlorocyclobutenone 192
using these conditions and were forced to abandon this approach.
SIMe3
OEt
Zn
Zn
TMEDA
Et:OH
O
'C
-rt16
h 43% l( 61
Meli, TMSCI
Et .,
98%
190
ciCCOCI
M 3SI
0
|20
OEt
EtO
191
CI
M03SI
(61)
EtO
193
192
Aor
hv
Me3Si
C
Eto
194
Thomas Lee of our laboratory has recently found that a-diazo thiol esters rearrange
in the presence of catalytic rhodium acetate at 25-80 C to generate (arylthio)ketenes in high
yield (eq 62). The facility and efficiency of this Wolff-type rearrangement is attributable to
the formation of a. 3-membered sulfonium ylide intermediate 196 from an initial carbenoid
species.1 17 Remarkably, when this reaction is conducted in the presence of unactivated
olefins such 198, cyclobutanones are obtained in outstanding yields (no cyclopropane
products are detected) (eq 63).
115.
116.
117.
(a) Danheiser, R. L.; Savariar, S. Tetrahedron Lett. 1987, 28, 3299. (b) Danheiser, R. L.; Selvaraj, S.;
Cha, D. D. Organic Synthesis 1989, 68, 32.
Ammann, A. A.; Rey, M.; Dreiding, A. S. Helv. Chim. Acta 1987, 70, 321.
(a) Hixson, S. S.; Hixson, S. H. J. Org. Chem. 1972, 37, 1279. (b) Boyer, S. K.; Edwards, B. J. Org.
Chem. 1980, 45, 1686.
91
A2
RN
h2(OAc)4
ArS
ArS
|
+
H
ArS
197
196
196
195
(62)
C97
O
0.02 equiv Rh 2 (OAc) 4
0
+ A,N
+ ArS
2
CICHCCIl
A
3.5 h
SAr
92%
199
195
198
(63)
We believed that this chemistry could provide an important new route to the 2trialkylsilylcyclobutenones
needed for the generation of TAS-vinylketenes via electrocyclic
ring opening. Our strategy is illustrated below (eq 64).
SiMe 3
0
ArSN
CH2 CI 2
+
195
Rh(OAc) 4
Me3Si
0
Me3Si
reduction
M
0
\y
(64)
EtO
OEt
SAr
EtO
201
200
191
Aor
hv
Me 3Si
C
EtO
194
To date, the [2 + 2] cycloaddition of the (arylthio)ketenes with trialkysilylsubstituted acetylenes has produced only a mixture of complex products under the standard
conditions we have successfully employed for cycloadditions with unactivated alkenes.
When no catalyst is present, no reaction occurs between 195 and 191. Exposure of the
(trialkysilyl)acetylene
191 to the rhodium catalyst in the absence of 195 does not lead to
alkyne decomposition.
92
In contrast to the silyl-substituted alkoxyacetylene
191, methoxypropyne
201118
reacted with the (arylthio)ketene generated from 195 to afford the desired cyclobutenone in
92% yield (eq 65).. A competition experiment was then conducted to determine the relative
efficiency of this alkyne (201) and cyclopentadiene as traps for the (arylthio)ketene.
The
alkyne in fact proved to be a very effective trap producing cyclobutenone 202 in 79%
yield. It thus appears that the complications arising in the [2 + 2] cycloaddition of the
(arylthio)ketene
and the silylalkyne 191 are due to the trialkylsilyl group attached to the
alkyne.
Me
Rh2(OA) 4
O
h-=406
ArS
'195
Me
92%
OMe
MeO
SAr
202
201
2) TAS-Vinylketenes via Isomerization of Trialkylsiloxyalkynes
As discussed earlier (p. 55), siloxyalkynes with small to medium-sized silyl groups
(Me3Si, Et3 Si, t-BuMe2Si) undergo a facile salt-promoted isomerization to form
silylketenes. Recently, Julia has reported that lithium acetylides react with lithium tert-butyl
peroxide to produce lithium ynolates in good yield (eq 66).119 We expected that the
application of the Julia protocol to enynes might provide a very expeditious route to
conjugated siloxyalkynes that upon warming to 25 °C might isomerize to produce TASvinylketenes.
118.
119.
The author would like to thank Michael D. Lawlor for the preparation of methoxypropyne (210).
Julia, M.; Saint-Jalmes, V. P.; Vereaux, J.-N. Synlett. 1993, 233.
93
-70 -- 0 °C
THF
Ph
L
203
+
BuOOLi
Ph
-
+
OLi
t-BuOLi
206
205
204
(66)
(i-Pr) 3 SiCI
-70 °C
60%
Ph
OSi(-Pr)
3
207
Initially, model studies were conducted using commercially available phenylacetylene
(208; eq 67). (Trimethylsilyl)phenylketene 211 was produced using this method in 16%
yield.
n-BuLi
THF
Ph
-
t-BuOOLi
THF
-78 -78 OC
2h
h
208
(67)
O LII
Ph
209
addition
210
16%
SiMe 3
211
Inverse addition of the lithium ynolate to a solution of excess trialkysilyl chloride
was found to be necessary in order to isolate the silylketene. If the addition was reversed
and the trialkysilyl chloride was added to the ynolate solution, the t-butoxide generated as a
byproduct during the reaction added to the silylketene producing the ester 212.
O
Ph
Ot-Bu
SiMe 3
212
When bulkier silyl chlorides such as t-BuMe2SiCl and Et 3 SiCl were used to quench
the ynolate, formation of the desired acetylene was initially observed by IR. Upon stirring
at room temperature (5-10 h), however, IR analysis revealed that isomerization to the
94
ketene did not occur and the acetylene decomposed. Further work will be necessary to
determine whether this strategy can provide a useful route to silylvinylketenes.
Summary
The photochemical Wolff rearrangement of -silyldiazo ketones is an excellent
method for the preparation of a variety of substituted TAS-vinylketenes.
We believe that in
principle some of the other strategies presented above will also provide access to
substituted TAS-vinylketene derivatives; however, further investigation is necessary.
95
CHAPTER 4
APPLICATIONS OF TAS-VINYLKETENES
AS DIENES IN THE DIELS-ALDER REACTION
The Diels-Alder reaction is a very powerful and versatile method for the
construction of six-membered rings.12 0 The use of highly substituted dienes has greatly
expanded the utility of this reaction as applied to the synthesis of more richly functionalized
rings. 12 1 The desired oxidation level at each carbon of the newly formed ring can often be
obtained by choice of the appropriately functionalized diene. Cyclohexenones are valuable
synthetic intermediates as well as attractive targets for assembly via the Diels-Alder
strategy. For example, Danishefsky has developed a particularly elegant approach to
achieve the synthesis 4-substituted cyclohexenones. 12 2 In general, this transformation
employs an oxygenated diene where X and Z are alkoxy substituents which confer high
reactivity and regiospecificity on the diene, and W is an electron withdrawing group such
as a carbonyl (eq 68). An example of the cycloaddition of "Danishefsky's diene" (trans-l1,3-butadiene) with methacrolein is shown below (eq 69).
methoxy-3-trimethylsilyloxy-
+
Ll
F)
*-
(68)
Z
benzene
80 °C 20 h
Me3SiOX
213
OM
(69)
CHO
213 OMe
120.
121.
122.
For reviews, see: (a) Oppolzer, W. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: New York, 1990, Vol. 5, pp 315-399. (b) Carruthers, W. In Cycloaddition Reactions in
Organic Synthesis; Pergamon Press: New York, 1990, pp 1-208. (c) Wollweber, H. InMethoden der
organischen Chemie (Houben Weyl); Mueller, E. Ed.; Georg Thieme; Stuttgart, 1970; Vol. 5/1c, pp. 9771210. (d) Fringuelli, F.; Taticchi, A. Dienes in the Diels-Alder Reaction; John Wiley & Sons; New York,
1990.
For some examples of cycloadditions of heterosubstituted dienes, see: (a) Petrzilka, M.; Grayson, J. I.
Synthesis, 1981, 753. (b) Danishefsky, S. Acc. Chem. Res. 1981, 14, 400.
Danishefsky, S.; Kitahara, T. J. Am. Chem. Soc. 1974, 96, 7807. (b) Danishefsky, S.; Kitahara, T. J.
Org. Chem. 1975, 40, 538.
96
Trost has also applied the Diels-Alder strategy for the synthesis of cyclohexenone
ring systems.12 3
In contrast to cycloadditions based on Danishefsky's diene, Trost's
reaction leads to the formation of substituted cyclohexenones in which the electronwithdrawing group is located at the C-5 position of the new ring (eq 70). A typical
example of Trost's cyclohexenone synthesis is shown in equation 71.
RO f5
1r
ArS
W
H or OH
W
(70)
1) BF3 OEt2
EtQO
95%
AcO>
+
CO0
2 Me
X
215
ArS
OW
r
2) Mg(OCH3) 2
MeOH, rt
71%
Pb(OAc)4
2I g(OCH benzene
C2OMpe 80°C
o
10 min
ArS
Of.CO 2Me
(71)
AcO
ArS
215
214
m-CPBA
CH 2CI2
-78 -- 40 °C
O
A
Ac
C
2
Me
216
216
In principle, cyclohexenones bearing the electron-withdrawing substituent at C-6 of
the new ring should also be accessible via a Diels-Alder strategy. This approach requires
the cycloaddition of an electron-deficient olefin with a vinylketene.
Unfortunately,
vinylketenes are not effective dienes in [4 + 2] cycloadditions because they readily undergo
[2 + 2] cycloadditions with olefins as predicted by frontier molecular orbital theory (eq 72).
An objective of current interest is the development of a vinylketene equivalent which is
capable of participating in [4 +2] Diels-Alder reactions.
123.
Trost, B. M.; Vladuchick, W. C.; Bridges, A. J. J. Am. Chem. Soc. 1980, 102, 3554.
97
[4+ 2]
~[2
W
W
+ 2]
]
c/W+
(72)
Previous Vinylketene Equivalents
Considerable effort has been directed toward the development of stable vinylketene
equivalents in which the critical ketene functionality is masked so that the desired [4 + 2]
Diels-Alder reaction can occur. Both vinylketene acetals 24 and thioacetals12 5 have been
reported to undergo [4 + 2] cycloadditions in good yield. However, due to the limited
reactivity of these hindered Z-dienes, only highly electrophilic dienophiles react. Brassard
has condensed vinylketene acetals with 2-halogeno-1,4-napthoquinones in order to
synthesize a number of naturally occuring anthraquinones. 12 4 b-g For example, 2-acetyl1,1-dimethoxy-3-methylbuta-1,3-diene
(218) has been added to the napthoquinone 217 to
produce 2-acetyl-1,6,8-trimethoxy-3-methylanthraquinone
(219) in 50% yield as a key
step in the synthesis of a coccid pigment (eq 73).
H3CO
HC
+
H3CO
OCH3
refluxing
xylenes
COCH3
50%
(73)
O
217
218
219
Carey has employed vinylketene thioacetals in [4 + 2] cycloadditions with highly
reactive dienophiles (Scheme 18). The expected cycloadducts can be converted to
cyclohexenones via a variety of hydrolytic methods. It is noteworthy that less reactive
124.
125.
For example of vinylketene acetals, see: (a) McElvain, S. M.; Morris, L. R. J. Am. Chem. Soc. 1952, 74,
2657. (b) Banville, J.; Grandmaison, J. L.; Lang, G.; Brassard, P. Can. J. Chem. 1974, 52, 80. (c)
Banville, J.; Brassard, P. J. Chem. Soc., Perkin Trans. 1 1976, 1852. (d) Banville, J.; Brassard, P. J.
Org. Chem. 1976, 41, 3018. (e) Grandmaison, J. L.; Brassard, P. Tetrahedron 1977, 33, 2047. (f)
Grandmaison, J. L.; Brassard, P. J. Org. Chem. 1978, 43, 1435. (g) Roberge, G.; Brassard, P. J. Chem.
Soc., Perkin Trans. 1 1978, 1041. (h) Gompper, R.; Sobotta, R. Tetrahedron Lett. 1979, 921.
For some examples of vinylketene thioacetals, see (a) Carey, F. A.; Court, A. S. J. Org. Chem. 1972, 37,
4474.
(b) Kelly, T. R.; Goemer,
R. N.; Gillard, J. W.; Prazak, B. K. Tetrahedron
98
Lett. 1976, 43, 3869.
dienophiles such as diethyl maleate, diphenylacetylene, and p-benzoquinone did not react
with the unsaturated thioacetals.
Scheme
18
xylene
A 3h
HgCI2
MeOH
65 °C 15h
60%
90%
0
o
220
[
222
221
223
I
CH3 0H
H2S04
58%yield
from222
225
224
Activated vinylketene acetals and thioacetals have been prepared with electron
donating groups at the C-3 position of the diene to help promote reaction with weaker
dienophiles.
Brassard and coworkers first prepared dienes activated by alkoxy and
silyloxy groups positioned at C-3 and studied their reaction with naphthoquinones.12 4 c,fg
Similarily, Danishefsky has reported the Diels-Alder reaction of the highly nucleophilic
diene 226 with several dienophiles (eq 74).
OH
C
Me 3 SiO
CO2 Me
benzene
80 °C 30 min
OMe
89%
(74)
HO
C0 2 Me
CO2Me
226
228
227
Recently, other heteroatoms such as sulfur and nitrogen have been positioned at C-
3 to help activate dienes toward cycloaddition (eq 75 and 76).126 These reactions proceed
at much lower temperatures than the corresponding reactions with the C-3 methyl
12.6.
Konopelski, J. P.; Kasar, R. A. Tetrahedron Lett. 1993, 34, 4587.
99
substituted diene which requires 110 °C in order for the cycloaddition to occur. No
reactions with other less reactive dienophiles have been reported.
0
CO2 Me
CH3CI
rt 74h
68%
+Me
MeO 2 C
0
72zMCOMe
(75)
PhS
COMe
0
0_'
00
°3
N-Me
-Me (76)
rt 12h
86%
O
Due to the strong nucleophilic character of vinylketene thioacetals, these dienes
were found to participate in a Michael-type stepwise process (eq 77).127 As illustrated
below, diene 229 adds to quinone 230 via a 1,4-Michael addition, followed by the
elimination of methanol and tautomerization to produce 231. It is believed that the
nucleophilic diene reacts via its "s-trans" conformer, and therefore the Diels-Alder type
cyclization is not competitive with ring closure to the furan ring.
0
MeO
rt 3h
S
+
62%
(77)
MeO
S
229
127.
S
230
__V.
Danishefsky, S.; McKee, R.; Singh, R. K. J. Org. Chem. 1976, 41, 2934.
100
Vinylketenimines such as 232 have been demonstrated to serve as useful
vinylketene equivalents.93 , 128 These dienes are isolable by distillation, stable at room
temperature, and undergo [4 + 2] reactions to produce cycloadducts in moderate to good
yields (eq 78). Diels-Alder reactions of (silyl)vinylketenimines are even reported to
proceed in good yield with moderately reactive dienophiles such as methyl acrylate and
crotonate.
93
N-To
CH3CN
0
N 0
0
2000
(78)
95%
232
Tol
221
o
L
U j
233
234
o
Cub
[4 + 2] Cycloadditions with TAS-Vinylketenes
Previous work carried out in our laboratory had established that TMS-vinylketene
can function as a diene component in Diels-Alder [4 + 2] cycloadditions.
The scope of
these reactions, however, is restricted by the limited thermal stability of this vinylketene
which must be handled and stored in solution.
As discussed in Chapter 3, the
photochemical Wolff rearrangement provided us with access to a variety of substituted
TAS-vinylketenes. These substituted derivatives were found to be more robust than the
parent TMS-vinylketene, and are even stable to silica gel purification and heating in C 6 D6 at
80 °C for several days. We therefore undertook a systematic investigation of the utility of
substituted TAS-vinylketenes as four-carbon diene components in [4 + 2] cycloadditions.
Our objectives in this study included examining:
(1) the scope of the reaction with respect
to various types of TAS-vinylketene substitution; (2) the scope of the cycloaddition with
regard to the degree of activation required on the dienophile;
(3) the regiochemical course
of the [4 + 2] cycloaddition; and (4) the stereochemical course of the reaction.
128.
Sonveaux, E.; Ghosez, L. J. Am. Chem. Soc. 1973, 95, 5417.
101
(1) Reactions with Dimethyl Acetylenedicarboxylate
Cycloadditions with the highly reactive dienophile dimethyl acetylenedicarboxylate
(DMAD) were examined first. The optimal procedure for the Diels-Alder reaction of TASvinylketenes and DMAD involves heating a 0.9 M degassed toluene solution of silylketene
and 1.0-1.5 equiv of DMAD in a sealed tube at 150 C until tlc analysis indicates that
consumption of the silylketene is complete. At this point, the reaction is transferred to a
The final product is isolated by column
round-bottomed flask and concentrated.
chromatography.
The reaction of silylketene 175 with 1.0 equiv of DMAD in toluene at 150 °C for
24 h produced the expected phenol 236 in 98% yield after purification by column
chromatography (eq 79). Other conditions investigated for this transformation included:
C), elevating the reaction
conducting the reaction at higher temperatures (180-200
temperature for lengths of time, and using a larger excess (2.0-2.5 equiv) of the dienophile.
However, none of these variations afforded the desired product in as high a yield as the
optimal conditions described above. It is believed that the Diels-Alder cycloadduct 235
forms initially and then undergoes a facile tautomerization to afford the more stable phenol.
(iAPr)S
C
COiCH3
0
10DC 16 h
toluene
OH
COCH
2
(i-Pr)3Si
3
~~~~~~~~~~~~98%
Although
235
227
1H
~(79) j
CO2CH3
CO2CH3
C2
175
I
CO
2 CH3
(i-Pr)3 S
236
NMR spectral data supports the structural assignment of this
compound and includes a resonance at 11.57 ppm due to the proton of the hydroxyl group,
this proton was not exchangeable with D2 0. Also, no O-H stretching frequency of the
hydroxyl group was observed in the IR spectrum. Therefore, in order to confirm the
identity of phenol 236, the compound was acetylated according to the procedure of Hofle
102
and Steglich (eq 80).129 The 1 H NMR resonance at 11.57 ppm was not observed in the
acetylated product 237, and a new singlet appeared at 2.40 ppm corresponding to the three
protons of the new acetate methyl group. Apparently, in phenol 236 a strong hydrogen
bond exists between the hydrogen of the hydroxyl group and the carbonyl oxygen of the
ester which affects both the IR stretching frequency of the O-H bond and the ability of the
proton to undergo exchange with D 2 0.
The presence of the trialkylsilyl group in the Diels-Alder adducts should facilitate
further synthetic elaboration of these compounds.
We have confirmed that
protodesilylation of this type of phenol is a smooth process as evidenced by the production
of' 238 in good yield (eq 81). In the 1H NMR spectrum of desilylated 238, the resonances
in the 1.0-1.6 ppm region corresponding to the triisopropylsilyl group were not observed,
and a new singlet appeared at 6.84 ppm due to the aromatic proton at C-4.
(CH3CO) 2 0
DMAP, CH2CI2
rt, 18 h,
OH
(i-Pr)3 S
(-Pr) 3Si
OCOCH3
C02C
l
67%
(80)
CO2 CH3
C3 2CH 3
237
COCH
2
236
3
OH
20 eq TFA
CH 2C122 h _2
CO2CH(81)
80%
0CO
2 CH 3
238
Cycloadditions were also attempted with the less stable (triethylsilyl)-substituted
vinylketene 176 and acetylene dimethyl dicarboxylate (eq 82). Interestingly, this reaction
did not afford the expected phenol 239, but instead produced a mixture of two compounds
240 and 241 as characterized
by NMR and IR analysis (for a further discussion of this
reaction, see p. 116). The 1 H NMR spectra of 240 and 241 contain singlets at 6.72 and
129.
Hfle, G.; Steglich, W. Synthesis 1972, 619.
103
6.80 ppm respectively corresponding to the C-5 aromatic protons. A phenolic proton was
observed at 10.8 ppm for 241 but was not found in the spectrum of 240.
Et3 S
CO2CH3
17 C
+
OH
Et3 Si
150 °C, 16h
toluene
IHI
CO2CH3
(82)
98%
C02 CH3
CO2CH3
176
227
239
OSiEt3
-
COCH
2
OH
CO2CH 3
3
CO2 CH3
C4
240
29%
CH3
241
20%
The Diels-Alder reaction of DMAD and two other (triisopropylsilyl)vinylketenes
was explored (Scheme 19). Optimal conditions for these reactions included heating the
ketene with 1.0-1.5 equiv of DMAD at 150 °C in a solution of toluene contained in a sealed
tube. Tert-butyldimethylsilyl-substituted
ketene 183 (p. 87) also undergoes cycloaddition
with DMAD in 61% yield (not optimized).
Scheme 19
+
II
toluene
150°C
3h
Me
(i-Pr)3 Si
61%
2Me
COMe
CH2CH 2 Ph
179
(/'Pr3)Si
CH 2CH 2Ph
227
C"0
+
II
242
2Me
toluene
150 °C 4h
73%
D2 Me
CO2Me
181
227
243
104
(2) Reactions with Ethyl Cyanoacrylate
The reaction of TAS-vinylketenes with ethyl cyanoacrylate was next examined in
order to investigate the regiochemical course of the cycloaddition.
The regiochemical course of the cycloaddition of TAS-vinylketenes with
dienophiles where W is an electron-withdrawing
group is controlled by the directing effect
of the carbonyl group as shown below. The carbonyl group of the ketene is also expected
to donate enough electron density to the diene system via resonance to allow Diels-Alder
reaction to occur with electron poor dienophiles. The trimethylsilyl substituent is known to
exert only a weak directing effect on the Diels-Alder reactions of 1- and 2-(trimethylsilyl)
1.,3-dienes.
13 0
_
l
W
R3 Si
-
-1.,
+
W
[Fo
The cycloaddition of substituted TAS-vinylketenes with dienophiles where W is an
electron-withdrawing group can produce two diastereomers as a result of the endo and exo
orientations of the dienophile substituents in the transition state (Scheme 20). The Alder
endo rule states that the endo mode of addition is usually preferred when an unsaturated
substituent is present on the dienophile. The preference for the endo transition state is the
result of secondary orbital interactions between the dienophile substituent and the
electrons of the diene.
130.
(a) Fleming, I.; Percival, A. J. Chem. Soc., Chem. Comm. 1976, 681. (b) Fleming, I.; Percival, A. J.
Chem. Soc., Chem. Comm. 1978, 178. (c) Jung, M. E.; Gaede, B. Tetrahedron 1979, 35, 621. (d) Batt,
D. G.; Ganem, B. Tetrahedron Lett. 1978, 3323.
105
Scheme 20
R3SI
R3SI
endo
.,, W
w:,
R3Si
W
I
exo
/',
/'IW
In the case of the dienophile ethyl cyanoacrylate both substituents contain
unsaturation. Only one example in the literature has been cited concerning the endoselectivity of ethyl cyanoacrylate. 13 1 Thus, Mellor and coworkers have investigated the
Diels-Alder reaction of cyclopentadiene with a variety of xa,3-unsaturated nitriles and found
that the nitrile group affords low selectivity for the endo product (eq 83 and 84). This is
believed to be partly a result of the centrosymmetric nature of the nitrile group which
presumably leads to an unfavorable geometry for secondary overlap. It is also thought that
substitution at the position alpha to the nitrile group in the dienophile can introduce
repulsive interactions in the transition states of both the exo- and endo-addition, but that the
interaction is greater in the exo-transition state.
cyano exo
cyano endo
I
0
Me
CN
_ products
products
xylene
138 °C
86%
CN (83)
Me
CN
0
MeO2C
CN
Me
C17:83
benzene
40 °C
50%/a
A2CO2 Me
CN
CO2Me
25:75
131.
Mellor, J. M.; Webb, C. F. I. J. Chem. Soc. Perkin 1 1974, 17.
106
CN
(84)
The reaction of silylvinylketene 175 with 1.2 equivalents of ethyl cyanoacrylate13 2
in toluene at room temperature produced a 2:1 mixture of two cycloadducts 244 and 245,
both which exhibited the expected regiochemistry (eq 85). The two diastereomers were
separated by column chromatography and the structures of the major and minor products
were established based on 1 H NMR spectral analysis as discussed below. The rate of
reaction of silylvinylketene 175 with ethyl cyanoacrylate was much faster than with
DMAD. Cycloaddition occurs with ethyl cyanoacrylate at room temperature, whereas
reaction with DMAD requires heating at 150 °C.
(i-Pr)3Si
C
CO2 Et
(i-Pr) 3S
(i-Pr)3S
toluene
rt 24 h
3
(85)
+
98%
aS~C
(i-Pr
244
175
245
The 1 H NMR coupling constant data for the major and minor isomers is shown
below (Figure 4). The spectrum for the major diastereomer 244 exhibits a large geminal
coupling constant (Jab = 13.9 Hz) between the protons Ha and Hb at C-5. Protons Ha and
Hb also couple to the C-4 proton Hc, and have coupling constant values of similar
magnitudes Jac = 6.4 Hz and Jbc = 6.0 Hz. These values are consistent with typical
equatorial-equatorial and equatorial-axial values. The proton Hc appears as a multiplet and
is expected to be coupled to the protons of the methyl group attached at C-4 and Ha and Hb.
132.
The author would like to thank the Loctite Corporation for a generous supply of ethyl cyanoacrylate.
107
Figure 4
Et
(
ec
For the major diastereomer 244:
ab 13.9 Hz
J = 6.0 Hz
Jac = 6.4 Hz
For the minor diastereomer 245:
ab= 14.2
5.4
Ja
=
Hz
Hz
10.7 HZ
The spectrum of the minor diastereomer shows a large geminal coupling constant
(Jab = 14.2 Hz) between Ha and Hb, and the proton Hb is coupled to the proton Hc with Jbc
= 5.4 Hz. A large axial-axial coupling constant (Jac = 10.7 Hz) is observed between
protons Ha and HI-l. In this case, Hc is observed as a doublet of doublet of quartets at 2.81
ppm with coupling to both Ha and Hb as well as to the protons of the methyl group attached
at C-4.
For each diasteromer two half-chair conformations are possible (Scheme 21). For
the CN endo-product, it is difficult to predict which conformer is favored. However, for
the CN exo-product,
245a is predicted to be the more stable conformer with the C-4
methyl group and C-6 ethyl ester both positioned equatorially, since the alternative
conformer 245b has a very unfavorable 1,3 diaxial interaction between the methyl group
and ethyl ester. The large axial-axial coupling constant (Jac = 10.7 Hz) between Ha and Hc
observed in the
1H
NMR of the minor isomer indicates that the C-4 methyl group must
have an equatorial orientation. This is consistent with the conformation predicted for the
CN exo-product 245a.
The conformer 244a with the C-4 methyl group axial must
correspond to the major diastereomer, and the 1H NMR data is consistent with this
assignment. Therefore, 2:1 selectivity for the CN endo-product is observed in the reaction
between ethyl cyanoacrylate and silylketene 175.
108
Scheme 21
Si(iPr)
CN endo-product
(major diastereomer)
H/
0
O
H
r)3
3
OEt
2
Me
0
Hb
244a CN
244a
CN exo-product
(minor diastereomer)
I
.t
Hc
Interestingly, in the
u
CN
nb
245b
245a
H NMR spectrum the two products 244 and 245 the
methylene protons of the ethyl ester are diastereotopic and each proton appears as a doublet
of quartets in the 4.10-4.40 ppm region. The IR spectrum for 244 shows a stretching
frequency at 2240 cm- 1 due to the nitrile group as well as two carbonyl stretching
frequencies at 1690 and 1750 cm - 1 corresponding to the oc,-unsaturated
ketone and ester
carbonyl groups respectively. The IR spectrum of 243 likewise shows two carbonyl
stretching frequencies at 1695 and 1755 cm-1. The nitrile stretching frequency is not
observed for this diastereomer, however, this is not surprising since nitriles are typically
characterized by weak to medium absorptions in the triple-bond stretching region of the IR
spectrum and their intensity is known to decrease when electron-attracting atoms are
attached. 13 3
The structural formula for each of these diastereomers was determined
separately by elemental analysis to be C2 1 H3 50 3 SiN.
Ethyl cyanoacrylate also undergoes [4 + 2] cycloaddition with other substituted
TAS-vinylketenes
such as the 13-ionone derived ketene 177 (eq 86). The overall yield of
this reaction is 43% (86% yield based on recovered starting material). A 4:1 mixture of
diastereomers
246a,b
was obtained and could not be separated by column
chromatography. Polymerization of the dienophile ethyl cyanoacrylate appears to be a
133.
Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric Identification of Organic Compounds;
Fourth Edition, John Wiley & Sons: New York, 1981, pp 129-130.
109
significant problem in this reaction. The cycloaddition of silylketene 177 with the
dienophile may be slow as a result of the bulky cyclohexenyl group, and consequently in
this case polymerization of the dienophile may be a more significant problem. The
consistency of the reaction mixture often becomes "jelly-like", and typically 30-40% of the
vinylketene starting material is recovered. Optimal conditions were achieved when 3.0
equivalents of the dienophile were added in small portions (0.5 equivalents) over the course
of 93 h.
0o
(i.Pr)3Si
C
o
NCO CO2Et
+
It
toluene
rt 93 h
43%
(86)
II,,
177
1H
246a,b
NMR data was obtained for the major diastereomer of this reaction from
examination of the spectrum of a mixture of the two diastereomers.
The 1 H NMR
coupling constant data for the major isomer is shown below (Figure 5). A large axial-axial
coupling constant (Jac = 11.7 Hz) was observed between Ha and Hc, and an equatorial-
axial coupling constant (Jbc = 5.0 Hz) was noted between Hb and Hc. A large geminal
coupling (Jab = 14.0 Hz) is also seen between Ha and Hb.
Figure 5
"CN
Forthemajordiastereomer
246:
Jab= 14.0 Hz
Jbc = 5.0 Hz
Jac = 11.7 Hz
110
Two half-chair conformations exist for each of the diastereomeric products formed
in this cycloaddition (Scheme 22). For the CN exo-product the conformer 246b' is
predicted to be favored since the bulky C-4 substituent and C-6 ethyl ester are both
equatorial. In the alternative conformer 246b", an unfavorable 1,3 diaxial interaction
exists between these C-4 and C-6 substituents. For the CN endo-product, conformer
246a" is expected to be favored in which the bulky C-4 group is equatorial. Conformer
246a' has an unfavorable 1,3 diaxial interaction between the bulky group at C-4 and the
C-6 nitrile. Consequently, for this reaction it was not possible to determine which
diastereomer is the major product formed (CN exo or CN endo), since for both
diastereomers the most stable conformation should have the bulky C-4 group positioned
equatorial.
Scheme 22
Si(i-Pr)3
CN endo-
H
product He6
I
.
0
2 Et
iI
R
Hb
CN
246a'
246a"
CO2Et Sl(.pr)
R
SI(iPr) 3
CN exoproduct
He
/-=0
COEN
RCOE
Hb
H,
He
CN
CN
246b'
Hb
246b"
(3) Reactions with Nitroethylene and Nitropropene
TAS-vinylketenes also undergo cycloaddition with nitro olefins, a class of
compounds known to be excellent dienophiles that react under mild conditions. The
electron-withdrawing nitro group controls the regiochemistry of the cycloaddition.
111
Nitroethylene (248) was prepared using the procedure of Ranganathan 3 4 (eq 87).
Thus, commercially available 2-nitroethanol (247) was dehydrated using phthalic
anhydride, and nitroethylene (248) was collected in 25% yield (lit. 80%)134 via distillation
at reduced pressure. Nitroethylene can be stored at 0 °C as an 1.0 M solution in benzene.
o
HOO
N
HO-N2
0
k,
¢ o
2
247
,
A
25%
'N
2
0
N2
(87)
248
The reaction of silylketene 175 and 3.0 equivalents of nitroethylene (248) in
benzene at room temperature provided a 2:1 mixture of diastereomers 249a,b (eq 88).
The dienophile was added in three portions (one equivalent each) over 39 h to help reduce
polymerization that might occur when a high concentration of nitroethylene is present in the
reaction mixture. Two crystals of BHT were also added as a radical inhibitor to help
reduce polymerization.
(i.Pr)3Si
CO-0
NO2
BHT, benzene (iPr)3 S
rt 39h
(88)
85%
248
175
249a,b
By analysis of the spectrum of the diastereomeric mixture of 249a,b, 1H NMR
data was obtained for each individual diastereomer. The proton alpha to the nitro group
appears at 5.30-5.40 ppm.
mixture of 249a,b.
13 C
NMR and IR spectral data were also obtained for the
At this point, it was difficult to determine the stererochemical
assignment of the major and minor diastereomers.
This was due to the physical
inseparability of the two diastereomers and the poor resolution in the H NMR spectrum
134.
Ranganathan, D.; Rao, C. B.; Ranganathan, S.; Mehrotra, A.; Ivengar, R. J. Org. Chem. 1980,45, 1185.
112
from which it was impossible to extract necessary coupling constant data. In any case, due
to the high acidity of the doubly-activated a'-proton,
the initial product of this
cycloaddition probably readily undergoes equilibration, and so the identity of the major
isomer would not necessarily provide insight into the stereochemical course of the DielsAlder step.
In order to specifically investigate
the stereochemical
course of these
cycloadditions, we next investigated the cycloaddition of silylketene 175 and nitropropene
(250). As shown below, there are again two possible approaches (endo versus exo) for
the dienophile in the transition state of the cycloaddition. We believed that the endo product
would be the only adduct formed in this reaction, thus demonstating the endo selectivity of
this cycloaddition.
0 2N
(i-Pr) 3Sc
endo
('Pr) 3S
II
/
(i-PC x
O
('-Pr) 3S
2
Indeed, the reaction of silylketene 175 and 2.0 equivalents of nitropropene in
toluene at 110 °C for 116 h produced only the endo cycloadduct 251 as a white solid in
35% yield (eq 89). A single crystal of BHT was added to the reaction mixture to help
prevent polymerization. Elevated temperature was found to be necessary for this reaction:
when the reaction was conducted at room temperature for 23 h, the desired cyclohexenone
251 was produced in only 2% yield with 75% of the vinylketene 175 recovered.
113
(iPr)3 Si
Y
CO
+
(/'Pr3" 175
~ O
'~O
NO2
BHT, toluene
110 0 C 116h
35%
(i-Pr)
(89)
iP)
Htlun
250
251
Proof of the structure of 251 was established from 1H NMR coupling constant data
and an NOE study (Figures 6,7, and 8). The protons at C-5, Ha and Hb exhibit a strong
geminal coupling (Jab = 13.5 Hz). Proton Ha is assigned to be on the same side of the ring
as the nitro group (this assignment is confirmed in an NOE study discussed below) and it is
shifted downfield in the
1H
NMR spectrum.
Ha appears at 2.71 ppm as a doublet of
doublets, and a large axial-axial coupling constant (Jac = 9.8 Hz) is observed between Ha
and the C-4 proton Hc. Protons Hb and Hc, in contrast, show equatorial-axial coupling
(Jbc = 5.8 Hz). Conformer 251 is predicted to be favored since 251a has an unfavorable
1,3 diaxial interaction between the C-4 methyl and C-6 nitro groups. The 1 H NMR data
presented matches that predicted for the major conformer of the endo cycloadduct (Figure
7)1.
Figure 6
o
(iPr)3Si
.
O 2
H,
2.52 (im)
HC
Hb
2.71 (dd, J = 13.5,9.8 Hz)
2.27 (dd, J= 13.5,5.8)
-
251
Figure 7
H3
H,
D
Hb
251a
251
114
An NOE experiment was conducted to confirm the assignment of Ha and Hb
(Figure 8). Irradiation of the methyl group at C-6 led to enhancement at Hb and Hc. No
enhancement was observed for the resonance assigned as Ha, thus confirming the fact that
Ha is on the opposite side of the ring as the methyl group and on the same side of the ring
as the electron-withdrawing nitro group.
Figure 8
1
0oe0%
H3C
Irradiate
1
3%
8%
251
The IR spectrum for cyclohexenone 251 has a carbonyl stretching frequency at
1670 cm-1 and characteristic C-NO2 stretching frequencies at 1540, 1380, and 1350 cm- 1 .
The nitropropene used in the Diels Alder reaction described above was prepared on
a small scale (1.5 g of product) according to the procedure of Noland (eq 90).135
Commercially available 2-nitropropanol (252) was dehydrated using phthalic anhydride
and 2-nitropropene (250) was collected via distillation at reduced pressure as a faint green
liquid in 67% (lit. 57-72%)135 yield. Nitropropene can be stored neat at 20 °C as a low
melting solid for several days without decomposition.
0
HO
N02
252
135.
O
N02
67%
(90)
250
Miyashita, M.; Yanami, T.; Yoshikoshi, A. Organic Syntheses, 1990, Coil. Vol. 7, p 396.
115
(4) Reactions with Cyanoallene
Cyanoallene (255) was prepared via a known method136 from propargyl bromide
(eq 91). This reaction involves a complex of CuCN and KCN and initially affords
The isomerization of acetylene 254 to
propargyl cyanide (254) and cyanoallene.
cyanoallene is caused by the slightly basic KCN present in the reaction mixture.
Cyanoallene was produced in 25% yield (lit.
and can be stored neat under
90%)136
nitrogen at -20 °C for several months.
H
CuCN
H
=--
KCN, H20
CH2Br
H--
25%
253
+
CH 2 CN
(91)
C=C
CN
255
254
H
/
-CC
.CN
255
The reaction of silylketene 175 and 3.4 equivalents of cyanoallene (255) occurred
in toluene at 150 °C in a sealed tube to produce a white solid in 67% yield (eq 92).
Examination of the
1H
NMR spectrum revealed that the product was not the expected
phenol 257, but rather the isomeric silyl ether 256 (Figure 9). Thus, the 1H NMR
spectrum of 256 contained a resonance at 6.55 ppm and no phenolic proton in the 10-11
ppm region.
O
(/i.r) 3SlC
HCN
toluene
+ II
150C 31h
C
175
136.
OSI(i-Pr)3
CN
(92)
67%
255
256
Brandsma, L.; Verkruijsse, H. D. Studies in Organic Chemistry 8: Synthesis of Acetylenes, Allenes, and
Cumulenes; Elsevier Scienfific: New York, 1981, pp 173-175.
116
Figure 9
(i-Pr)
16.55 ppm
257
expected product
256
product obtained
The structure of the compound resulting from the cycloaddition of silylketene 175
and cyanoallene was established using an NOE study. Irradiation of the resonance at 6.55
ppm resulted in enhancement of resonances corresponding to both the C-4 methyl group
and the triisopropylsilyloxy group at C-2 (Figure 10).
Figure 10
Irradate
_
0%
I
Ao/n
256
The cycloaddition of silylketene 175 and cyanoallene (255) is presumed to occur
through the intermediate tautomers 258, 259, and 260 (Scheme 23). a-Silyl carbonyl
compounds are known to rearrange to afford silyl enol ethers in good yield and this
reactions requires only moderate temperatures.13 7 Therefore, it is believed that the ox-silyl
carbonyl tautomer 260 readily undergoes a 1,3-silyl shift to produce 256 (for an earlier
example of a Diels-Alder reaction that gives an aryl silyl ether see p. 104).
137.
Colvin, E. Silicon in Organic Synthesis; Buttersworths: London, 1981, pp 33-36.
117
Scheme 23
,v
/l
175
. n
~'i/
(Pr)
(i-Pr~i
r
I
O
o
r
1 .
OH
(i-Pr)
S 3 iCN
CN
3
=
255
258
257
211259
1,3 silyl
1,3silylft
shift
;N
L-
260
1
erCN
oSI(kpr)
3
256
(5) Cycloadditions with Other Dienophiles
Unfortunately, we found that less reactive dienophiles such as N-phenyl maleimide
and chloroacrylonitrile did not undergo the desired [4 + 2] cycloaddition with TASvinylketenes. Reaction with each of these dienophiles led in low yields to complicated
mixtures of several products.
Overall, we have demonstrated that a variety of substituted TAS-vinylketenes
undergo [4 + 2] Diels Alder reactions with reactive olefinic and acetylenic dienophiles in
Consistent with the reactivity of the parent
good to excellent yields.
(trimethylsilyl)vinylketene
(112, p. 69), the reactivity of these substituted TAS-
vinylketenes compares favorably to previously reported vinylketene equivalents. The
ketene carbonyl group dominates in controlling the regiochemical course of these reaction.
Cycloadditions of TAS-vinylketenes with monosubstituted dienophiles result in the
formation of a 6-substitued cyclohexenones. Particularly noteworthy is the stereochemical
course of these cycloadditions:
the reaction with nitropropene produced the endo
cycloadduct as the only isomer.
Overall, we have demonstrated the use of TAS-
vinylketenes as interesting four-carbon dienes in [4 + 2] cycloadditions to afford
cyclohexenone and phenolic products.
118
At present, the main limitation of this methodology is the modest reactivity of the
TAS-vinylketenes.
In this connection it would be extremely interesting to investigate the
reactions of TAS-vinylketenes bearing electron-donating substituents (e.g. -OR, -OSiR 3 ) at
the C-3 carbon. As in the case of Danishefsky's diene, these vinylketenes should exhibit
greatly enhanced reactivity and should combine in good yield with less activated
dienophiles such as acrylate. Studies are planned to examine the synthesis and chemistry
of this new class of TAS-vinylketenes.
119
CHAPTER
TAS-VINYLKETENES
5
AS FOUR-CARBON COMPONENTS
IN NEW [4 + 1] ANNULATION REACTIONS
Introduction
Currently there exists a great deal of interest in the development of new methods for
the construction of the cyclopentenone ring system. This is due to the fact that many
biologically active natural products have this 5-membered ring moiety as a major structural
feature. For example, molecules which incorporate the cyclopentenone ring system include
(Figure 11) cis-jasmone (261), an important constituent of many perfumes; the rethrolones
(262), the alcohol components of the insecticidal pyrethrin esters; and prostaglandins such
as PGA2 (263) and A7 -PGA1 methyl ester (264), both which are promising anticancer
agents.
138
Figure
11
0
R
HO
rethrolones
262
cis-jasmone
261
CO2H
d
~ Am
Am
OH
OH
7
A -PGA Me Ester
PGA 2
263
138.
264
Noyori, R.; Suzuki, M. Science 1993, 259, 44.
120
Cyclopentenones
can be synthesizedl
39
by a variety of different methods such as
the Nazarov1 40 and related cationic cyclizations, the Pauson-Khand colbalt-mediated
cyclization of alkynes with olefins,14 1 intramolecular aldol condensations, the insertion of
vinyl carbenes into C-H bonds,14 2 and [3 + 2] coupling reactions. 143
There is also a
limited number of [4 + 1] cycloaddition strategies for the construction of the
cyclopentenone ring system.l4 4 As described in Part II Chapter 3, we have recently
developed methodology that provides efficient access to a variety of stable TASvinylketenes. We believed that these unusual molecules could function as four-carbon
components in a versatile and conceptually novel [4 + 1] annulation strategy to synthesize
cyclopentenones.
As illustrated below, we anticipated that it would be possible to add
"carbenoid reagents" (nucleophilic species bearing appropriate leaving groups) to TASvinylketenes so as to generate a dienolate intermediate which could then cyclize to form a
new five-membered ring.
I
0
-S--s
L
1
-R3
R
13
.R4
4
R
R
R2
139.
140.
For reviews of methods to synthesize cyclopentenones, see: (a) Ellison, R. A. Synthesis, 1973, 397. (b)
Piancatelli, G.; D'Auria, M.; D'Onofrio, F. Synthesis, 1994, 867.
(a) Denmark, S. E. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press:
New York, 1990, Vol. 5, p 751. (b) Habermas, K. L.; Denmark, S. E. In Organic Reactions;
Ed.; John Wiley & Sons: New York, 1994; Vol 45, pp 1-158.
141.
142.
143.
144.
Paquette, L. A.
(a) Schore, N. E.. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press:
New York, 1990; Vol. 5, p 1037. (b) Schore, N. E. In Organic Reactions; Paquette, L. A. Ed., John Wiley
& Sons: New York, 1991, Vol. 40, pp 1-90.
(a) Karpf, M.; Huguet, J.; Dreiding, A. S. Helv. Chim. Acta 1982, 65, 13. (b) Williamson, B. L.;
Tykwinski, R. R.; Stang, P. J. Am. Chem. 1994, 116, 93.
(a) Noyori, R.; Shimizu, F.; Fukuta, K.; Takaya, H.; Hayakawa, Y. J. Am. Chem. Soc. 1977, 99, 5196.
(b) Noyori, R.; Yokoyama, K.; Hayakawa, Y. J. Am. Chem. Soc. 1973, 95, 2722.
For some examples of [4 + 1] annulations to form cyclopentenones, see: Ogura, K.; Yamashita, M.;
Furukawa, S.; Suzuki, M.; Tsuchihashi, G. Tetrahedron Lett. 1975, 2767. (b) Ogura, K.; Yamashita, M.;
Tsuchihashi, G. Tetrahedron Lett. 1976, 759. (c) Johnson, B. F. G.; Lewis, J.; Thompson, D. J.
Tetrahedron Lett. 1974, 3789.
121
[4 + 1] Annulation with Diazo Compounds
The first goal of our study was to establish the feasibility of the [4 + 1] annulation
strategy and to identify "methylene transfer" reagents that can participate in the reaction. As
discussed in Part II Chapter 2 of this thesis, TMS-ketene (and other silylketenes) are
known to react with a variety of carbanionic species including organolithium and
organocerium compounds, alkoxystannanes, and stabilized phosphorous ylides. Based on
this precedent, we anticipated that both unstabilized and moderately stabilized carbanion
derivatives would add to TAS-vinylketenes in the desired fashion. Highly stabilized
carbanion derivatives, however, were expected to be unsuitable partners for our annulation
due to the possible intervention of internal proton transfer at an intermediate stage of the
reaction.
The TAS-vinylketene 175 was selected as a representative ketene to test the
feasibility of the proposed annulation. Initial studies focused on diazomethane as the
"carbenoid" annulation component. We found that the desired transformation proceeds
smoothly to afford the expected (triisopropylsilyl)cyclopentenone
system 265 in 96% yield
(eq 93). This reaction is remarkably facile, with tlc analysis indicating that the formation of
the 5-membered ring product is complete upon warming to -20 °C.
It was neccessary to cool the solution of ketene 175 at -120 °C while diazomethane
was added. When the reaction was conducted at a higher temperature such as -78 C, the
desired product was isolated in only 70% yield with a -30% recovery of silylketene 175.
The annulation was also conducted using an excess of diazomethane (5 equiv), and the
desired cyclopentenone was produced in 93% yield. Interestingly, no cyclobutanone
formation due to the ring expansion of a possible cyclopropanone intermediate (see
discussion of the mechanism, p.123) was observed.
122
O
2 equiv CH2 N2
Et2O-CH 2 CI2
(i-PrSi
-120 °C-+ rt 3h
,(
\
c
I Hb
(93)
96%
(i-Pr)3S
175 C
265
O
,
265
175
The structural assignment for cyclopentenone 265 is based on analogy with data
reported for similar ox-silylcyclopentenones. The carbonyl stretching frequency for
cyclopentenone 265 is observed in the IR spectrum at 1685 cm- 1 and is consistent with the
stretching frequency reported for similar cyclopentenones (Figure 12).145,146The
13 C
NMR spectrum for cyclopentenone 265 exhibits a carbonyl resonance at 213.3 ppm which
is very similar to the
13 C
silylcyclopentenones
such as 268 and 269 (Figure 13).146
NMR resonances reported for the carbonyl carbons in other a-
Figure 12
0
0
Me3 Si
(i-Pr)SI
266
265
1665 cm'1
1685 cm-'
(film)
(KBr)
0267SiMe
M03SI
3
Ph
267
268
1'
1700cm
(film)
1690 cm-'
(film)
Figure 13
I
0
268
212.5
Me 3 Si
265
|
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0
SiMe 3
213.3
Ph
267
|
| 211.4
|
Scheme 24 outlines several alternative pathways which could account for the
mechanistic course of this [4 + 1] annulation. The initial addition of diazomethane to the
145.
146.
Shih, C.; Fritzen, E. L.; Swenton, J. S. J. Org. Chem. 1980, 45, 4462.
(a) Sawada, H.; Webb, M.; Stoll, T.; Negishi, E. Tetrahedron Lett. 1986, 27, 775. (b) Negishi, E.;
Holmes, S. J.; Tour, J. M.; Miller, J. A.; Cederbaum, F. E.; Swanson, D. R.; Takahashi, T. J. Am. Chem.
Soc. 1989, 111, 3336.
123
vinylketene should be highly stereoselective due to the shielding effect of the bulky silyl
group 7 4 and should result in the formation of the Z enolate 270.
Scheme 24
Enolate 270 may undergo internal displacement of nitrogen and produce the fivemembered ring system directly. However, the planar structure of the dienolate system
probably makes it difficult for 270 to achieve an arrangement in which the ,t electrons are
suitably situated for a direct backside displacement of nitrogen. A second possible pathway
involves ionization of the enolate 270 to give the pentadienyl cation 272 which can
undergo 4
electrocyclic ring closure to the cyclopentenone ring. This type of 4t
electrocyclic closure of a pentadienyl cation is well known in synthetic methods such as the
Nazarov cyclization
14 0
and the cyclization of a species generated via the in situ epoxidation
of vinylallenes (Scheme 25).147 Mechanistically, this stereoselective process is rationalized
by the initial hydroxyl-directed formation of allene oxide 273, which subsequently
isomerizes to the oxidopentadienyl cation 274 or vinylcyclopropanone 275. Conrotatory
electrocyclic ring closure then affords the cyclopentenone ring 276.
147.
Kim, S. J.; Cha, J. K. Tetrahedron Lett. 1988, 29, 5613.
124
Scheme 25
R3
OH
R
R2
C En
t-BuOOH
-VO(acac) 2
0
4
R
H
CH
2CI
2, rt .
R2
R4
R1
273
I
O-
conrotatory
H
electrocyclic
ring closure
OH
and/or
R3
R1
276
274
275
This type of 4iTelectrocyclic closure of a pentadienyl cation is also proposed in the
biosynthetic pathway for prostanoid synthesis by marine organisms.'4 8 For example, the
biosynthesis of prostanoid preclavuone A from arachidonic acid involves the conversion of
an allene oxide intermediate to a cyclopentenone product (Scheme 26).
Scheme 26
,OOH
O
COOH
COOH
==r fl-C>C~~~~
S~H~1
O-
T
.
'
~
n.
1
preclavuoneA
CCH
n
«~n-C
5H 1,
b~~ (~/=H
n-C,H,,
'i .
-n-C
-
CHOOH
f CH ,,
A third alternative mechanism to account for our [4+1] annulation (Scheme 24)
involves the internal displacement of nitrogen to form the vinylcyclopropanone 271
followed by electrocyclic ring opening to generate pentadienyl cation 272. Electrocyclic
148.
For evidence for such a biosynthetic
pathway, see:
(a) Brash, A. R.; Bairtschi, S. W.; Ingram, C. D.;
Harris, T. M.; J. Biol. Chem. 1987, 262, 15829. (b) Baertschi, S. W.; Ingram, C. D.; Harris, T. M.;
Brash, A. R. Biochemistry,
1988, 27, 18. (c) Brash, A. R.; Baertschi, S. W.; Ingram, C. D.; Harris, T. M.
Proc. Natl. Acad. Sci. USA 1988, 85, 3382. (d) Corey, E. J.; d'Alarcao, M.; Matsuda, S. P. T.; Lansbury,
P. T.; Yamada, Y. J. Am. Chem. Soc. 1987, 109, 289.
(e) Corey, E. J.; Matsuda, S. P. T.; Nagata, R.;
Cleaver, M. B. Tetrahedron Lett. 1988, 29, 2555. (f) Hamberg, M.; Miersch, O.; Sembdner, G. Lipids,
1988, 23, 521. For a demonstration of the feasibility of this proposal, see: Corey, E. J.; Ritter, K.; Yus,
M.; Najera, C. Tetrahedron Lett. 1987, 28, 3547.
125
closure would then produce the cyclopentenone product. Recall, that Baukov has found
that diazomethane adds to TMS-ketene to produce the three-membered ring TMScyclopropanone in 50% yield.8 8 Cyclopropanones, because of their instability, are
generally only observed as transient intermediates and trapped with nucleophiles.14 9 The
proposed vinylcyclopropanone
271 is predicted to be very unstable and the C2-C3 bond is
expected to readily undergo cleavage to the oxallyl species.
An example of a
vinylcyclopropane which presumably undergoes C2-C3 bond cleavage to produce a
pentadienyl cation system which is then trapped by furan is illustrated in Scheme 27.150
Scheme 27
2
hv
-190 C
_
A
Currently, for our [4 + 1] annulation we favor the mechanistic pathway proceeding
vial ionization of the initial adduct 270 to form the oxidopentadienyl cation 272 followed
by electrocyclization. Obviously, further study is needed to elucidate the details of the
mechanism of the annulation and this problem will be the subject of future investigations in
our laboratory.
We next examined the [4 + 1] annulation using commercially available TMSdiazomethane (276),151a thermally stable diazoalkane often used in place of hazardous
149.
150.
151.
For a review of cyclopropanone chemistry, see: (a) Wasserman, H. H.; Clark, G. M.; Turley, P. C. Top.
Curr. Chem. 1974, 74, 73. (b) Wasserman, H. H.; Berdahl, D. R.; Lu, T.-J. In The Chemistry of the
Cyclopropyl Group, Rappoport, Z., Ed., John Wiley & Sons: New York, 1987, pp 1455-1532.
Barber, L. L.; Chapman, O. L.; Lassila, J. D. J. Am. Chem. Soc. 1969, 91, 3664.
(a) Seyferth, D.; Dow, A. W.; Menzel, H.; Flood, T. C. J. Am. Chem. Soc. 1968, 90, 1080. (b) Shioiri,
T.; Aoyama, T.; Mori, S. Organic Synthesis 1990, 68, 1. (c) Mori, S.; Sakai, I. Aoyama, T.; Shioiri, T.
Chem. Pharm. Bull. 1982, 30, 3380.
126
diazomethane. TMS-diazomethane reacts with silylketene 175 at room temperature to
afford cyclopentenone 277 in 84-92% yield. Desilylation of 277 with potassium
carbonate and methanol affords 265 (eq 94).
(i-Pr)
3
'O
1.6 equivTMSCHN2 (276)
O
I
O
277
265
184-92%
175
The bis-silylcyclopentenone 277 exhibits a characteristic carbonyl stretching
frequency
at 1665 cm- 1 in the IR spectum.
The substituents at C-4 and C-5 on the
cyclopentenone ring were assigned to be trans based on 1H NMR coupling constant data
reported for cis and trans-4,5-dimethyl-cyclopent-2-en-l-one
(278 and 279; Figure
15).152 For the cis stereoisomer 278, the C-5 proton (Hb) is reported to appear at 2.44
ppm as a doublet of quartets in the 1H NMR spectrum. The coupling constant between Ha
and Hb is 7.3 Hz, and the dihedral angle between the two protons is estimated to be 0 ° . On
the other hand, in the 1 H NMR spectrum for the trans stereoisomer 279, the C-5 proton
(Hb) appears at 1.86 ppm as a multiplet with a small coupling constant between Ha and Hb
of 2.5 Hz with an estimated dihedral angle of 1200 between these two protons. It should be
noted that the coupling constants observed for silylcyclopentenone 265 (eq 80, p. 123) are
also consistent with these values (Figure 14). In the 1H NMR spectrum for our bis-silyl
annulation product 277, a very small coupling constant (Jab = 0.80 Hz) is observed
between Ha and Hb which is consistent with the small coupling constant reported for trans-
4,5-dimethyl-cyclopent-2-en-1-one (279), and therefore, the stereochemistry of the C-4
and C-5 substituents on the ring is assigned to be trans. This trans stereochemistry agrees
with the mechanistic pathway we predict for this [4 + 1] annulation where ring closure
152.
Strike, D.; Smith, H. Ann. N. Y. Acad. Sci. 1971, 180, 91.
127
occurs via a 4xcelectrocyclic conrotatory process (for a more detailed discussion see p.
143).
Figure
14
Solvents have a significant effect on the rate of this [4 +1] annulation. When the
reaction is conducted in a non-polar solvent such as hexane, the desired cyclopentenone
277 is produced in 45% yield after 48 h at room temperature.
In contrast, when a very
polar solvent such as acetonitrile is employed, the cyclopentenone is generated in 68% yield
after only 22 h at room temperature. This result suggests that charge separation is involved
in the transition state of the reaction.
Stabilization of this charge occurs through
interactions with polar solvent molecules and results in an increase in the rate of the reaction
when it is conducted in a polar solvent.
A variety of other TAS-vinylketenes react with TMS-diazomethane to afford silyl
cyclopentenones in good yield (eq 95, 96, and 97). These reactions were conducted using
1.5 equiv of TMS-diazomethane
on a scale to give 130-140 mg of products which were
purified using column chromatography.
Note, that the annulation is not limited to
(triisopropylsilyl)ketenes; the somewhat less stable triethylsilyl derivative 176 also was
found to react in excellent yield (eq 95).
128
0~_O
Et3SI X
Me3SiCHN2
0
E3SI
rt21
"H
\
SiMo
'
(95)
95%/
176
(iPr)3 Si
c
280
o
Me3 SiCHN
2
(i-Pr)Si
CH C1
(96)
81%
Ph
182
281
{FPr)
3S11
(-Pr)
Me SiCHN2
3
S
I *0
C
~C8H2Cl
O
2
o
rt 29.5h
(97)
83%
174
282
The structural assignment for the trans stereoisomers 280 and 281 was based upon
1H
NMR coupling constant data. For both compounds the coupling constant observed
between Ha and Hb is relatively small (280, Jab = 1.9 Hz and 281, Jab = 2.1 Hz) which is
consistent with the data for trans-4,5-dimethyl-cyclopent-2-en-
-one 279152 discussed
earlier. The 1H NMR spectral data for the bicyclic enone 282 is in excellent agreement
with
data
reported
by
Negishil
4 6b
for the
related
compound
9-
(trimethylsilyl)bicyclo[4.3.0]non-1(9)-en-8-one (268; p. 123). The very small coupling
constant (Jab = .0 Hz) suggests that the trimethylsilyl group has the indicated exo
orientation.
Ketones generally do not react with TMS-diazomethane. However, it has been
found that in the presence of boron trifluoride etherate, ketones can be homologated. 53 It
is presumed that the BF3 .OEt2 coordinates to the oxygen atom of the ketone and helps
activate the carbonyl toward the nucleophilic addition of TMS-diazomethane.
153.
(a) Hashirnoto, N.; Aoyama, T.; Shioiri, T. Chem. Pharm. Bull. 1982, 30, 119. (b) Hashimoto, N.;
Aoyama, T.; Shioiri, T. Tetrahedron Lett. 1980, 21, 4619.
129
Consequently, we believed that Lewis acids such as BF3OEt2 might be effective at
accelerating our [4 + 1] annulation and studies were conducted accordingly (Table 11).
Table 11
I
I
cC
(i-Pr)3S
TMSCHN
2
CH2 CI2
iMe3
\c"
(iPr3 175
277
277
175
1
1
Entry
Conditions
1
1.5 equiv BF 3O-Et 2
-10 °C, 2 h
complex mixture of products
1.0 equiv BF 3OEt
37% yield
....
I
2
'li
T---
....
T----I--T----T
-78 °C to rt, 24 h
2
Results
liT'-----
3
0.05 equiv BF 3.OEt 2
0 C to rt, 4 h
43% yield;
ketene recovered in 43% yield
4
1.0 equiv TMSC1
0 °C to rt, 26 h
48% yield
5
0.05 equiv Me2AlCl
-78 °C
ketene decomposed upon exposure
to Lewis acid
1.5 equiv A1Cl3
ketene decomposed upon exposure
to Lewis acid
6
rt
7
8
0.5 equiv AlC13
-78 C to rt, 5 h
ketene recovered in 100% yield
1.5 equiv ZnC12
complex mixture of products;
ketene recovered in 39% yield
rt to 50 C, 4 h
9
0.5 equiv ZnC12
-78 C to rt, 24 h
no reaction observed by tlc
10
0.5 equiv TiC14
-78 °C
ketene decomposed upon exposure
to Lewis acid
Overall, Lewis acid catalysis of the addition of TMS-diazomethane to silylketene
175 was unsuccessful.
In a few cases (entries 2-4), the desired cyclopentenone 277 was
produced, however, no increase in the rate of formation of product was observed.
Unfortunately, in most cases decomposition of the silylketene was observed immediately
130
upon exposure to the Lewis acid (entries 5, 6, and 10) or a complex mixture of products
resulted upon the addition of TMS-diazomethane (entries 1 and 8).
Shioiri has reported the preparation of a number of a-trimethylsilyldiazoalkanes
via
the alkylation of the lithium salt of TMS-diazomethane and an alkyl halide as illustrated
below.
154
In principle, these reagents should be useful in our [4 + 1] annulation as
"alkylidene transfer" reagents to synthesize cyclopentenones bearing two substituents at the
C-5 position of the new 5-membered ring.
RX
+
Me3SiC(Li)N
Me 3SiC(N2)R
Initially, we attempted to generate a-trimethylsilyldiazoethane
(283) from the
alkylation of the lithium salt of TMS-diazomethane with methyl iodide and then add it
without isolation to a TAS-vinylketene.
Thus, methylation was conducted at -23 °C, the
silylketene 175 was then added at 0 °C, and the reaction mixture was allowed to warm to
room temperature (eq 98).
Unfortunately, upon work-up the silylketene 175 was
recovered in 100% yield, and it is believed that a-silyldiazoalkane
283 was not generated.
Note that Shioiri has not reported the synthesis of this specific a-silyldiazoalkane.
n-BuLi, Mel
THF
Me3SiCHN2
CH3
-78 -* -23 °C 4.5 h
MeSiC
M 3 SiC- N2
(98)
283
175
° IC - rt 24 h
0
(i-Pr) 3 S
15,4, Aoyama, T.; Shioiri, T. Tetrahedron Lett. 1980, 27, 2005.
131
iM
e
S
3
Shiori has reported the preparation of 284 in 77% yield. 15 4 Employing his
procedure, we alkylated lithium TMS-diazomethane (prepared from TMS-diazomethane
and n-butyllithium) with benzyl chloride and obtained 284 after vacuum distillation (140
°C, 0.25 mmHg) in 66% yield (eq 99).
n-BuLi, THF
PhCH 2 CI
66%
Me 3SiCHN2
276
Me 3SiC(N 2)CH 2Ph
284
(99)
Unfortunately, no significant reaction was observed between silylketene 175 and
a-silyldiazoalkane 284 in dichloromethane at temperatures ranging from 0 to 50 °C (entries
1--2;Table 12). Attempted catalysis of the reaction using BF 3 .OEt 2 only afforded unreacted
TAS-vinylketene and a complex mixture of products (entries 3-4).
Table 12
(PPr)3S
C-
284
CH2CI2
0
285
175
Entry
....
1
CH2 Ph
SiMe3
(i.Pr)sSi
Conditions
Results
1.5 equiv TMSC(N2)CH2Ph
no reaction observed by tlc
I
0 °C to rt, 3 d
2
1.3 equiv TMSC(N 2 )CH 2 Ph
rt, 24 h then 50 °C, 2 h
no reaction observed by tlc
3
0.05 equiv TMSC(N2)CH2Ph
0.05 equiv BF 3 .OEt 2
complex mixture of products;
ketene recovered in 50% yield
1.5 equiv TMSC(N2 )CH2 Ph
complex mixture of products
rt to 50 °C, 4 h
4
1.0 BF 3 OEt 2
-78 °C to rt, 24 h
The preceeding results led us to conclude that TMS-diazoalkanes are considerably
less reactive than diazoalkanes, and that substitution of bulky groups such as benzyl further
132
retard the rate of addition to vinylketenes. We therefore next turned our attention back to
simple diazoalkanes lacking the silyl substituent. 2-Diazopropane (289) was selected for
our initial studies and was prepared according to the literature procedure (eq 100).155
Thus, the acetone azine 287 was first prepared via the reaction of acetone and hydrazine
monohydrate. Azine 287 was then treated with anhydrous hydrazine (distilled from
refluxing hydrazine hydrate and sodium hydroxide pellets) to produce essentially pure
acetone hydrazone 288. Next, a solution of 288 in diethyl ether was treated with mercuric
oxide and potassium hydroxide in ethanol. 2-Diazopropane (289) was codistilled with
diethyl ether at reduced pressure into a cooled receiver to yield a bright orange -2.0 M
ethereal solution.
0
286
.NO
NH
NHH2
2
287
288
| HgO
KOH
>N2
289
The silylketene 175 was treated with several portions (5, 10, and 38 equiv) of 2diazopropane (eq 101). No significant reaction was observed by tlc analysis, and 73% of
the silylketene was eventually recovered. Two minor products were isolated, and ketene
carbonyl stretching frequencies were observed in the IR spectrum for both compounds
indicating that no addition of 2-diazopropane had occurred. Again, the steric bulk of this
reagent is thought to be an important factor preventing the addition of the diazo compound
to the ketene.
155.
(a) Day, A. C.; Whiting, M. C. Organic Syntheses, 1988, Coll. Vol. 6 10. (b) Andrews, S. D.; Day, A.
C.; Raymond, P.; Whiting, M. C. Organic Syntheses, 1988, Coll. Vol. 6, 392.
133
(i-Pr)3 S1
C
53 equiv (total)
CH2CI2
(-ROO
+ N28<
-78--0C
(101)
289
175
290
Several attempts were made to add commercially available ethyl diazoacetate (291)
to TAS-vinylketene
175 (Table 13). When silylketene 175 and ethyl diazoacetate were
stirred together at room temperature followed by elevated temperatures, no reaction
occurred (entry 1). Reaction in the presence of catalytic amounts of rhodium acetate also
gave none of the desired product, and in this case it appeared that cyclopropanation had
occurred at the C-3/C-4 double bond of the silylketene to afford a cyclopropyl silylketene in
-20% yield (entry 2). BF3.OEt 2 catalysis also did not promote formation of the desired
cyclopentenone and the TAS-vinylketene
was recovered in 45% yield (entry 4). The 1H
NMR spectrum of the minor product isolated in this reaction showed a quartet at 5.40 ppm
presumably corresponding to Ha, thus suggesting that no cyclization had occurred.
It is well known that ketones undergo smooth homologation using the lithium anion
of ethyl diazoacetate. 156 We therefore generated, the anion of ethyl diazoacetate using
lithium diisopropylamide and added it to a solution of the silylketene 175 in THF (entry 3).
Tlc analysis indicated that although several products had formed, there appeared to be no
cyc:lized product as evidenced by the quartet present in the 1H NMR spectrum presumably
corresponding to Ha.
156.
Nagao, K.; Chiba, M.; Kim, S.-W. Synthesis, 1983, 197.
134
Table 13
Entry
conditions
Resul ts
1
CH 2 C1 2
ketene recovered in 83% yield
0.02 equiv Rh 2 (OAc) 4 ,
CH 2C12, rt, 30 min
-20% C-3/C-4 double bond addition
1.0 equiv n-BuLi, THF
mixture of uncyclized products
rt to 40 °C, 29 h
2
3
ketene recovered in 42% yield
-78 °C to rt, 5 h
4
0.01 equiv BF 3.OEt 2
mixture of uncyclized products
ketene recovered in 45% yield
CH 2C12
-78 °C to rt, 20 h
[4 + 1] Annulation with a-Halocarbanions
(Chloromethyl)lithium, ClCH2 Li, has limited synthetic utility due to its extreme
thermal instability. However, this reagent has been generated and captured in nearly
quantitative yield by the addition of n-butyllithium or methyllithium to mixtures of
chloroiodomethane
chlorohydrins (eq 102).1 57
R"Li
R
0,>O
+ ICH2CI
R'
--
with aldehydes and ketones in THF at -78 °C to produce epoxides and
1
R
L
R'~
'CHCIJ
R
OH
>4
R
"CH 2 CI
><
or
R'
(102)
·
The success of this method depends on rate relationships between several fast
reactions. Halogen-metal exchange between chloroiodomethane and the organolithium
reagent must be faster than the addition of the organolithium reagent to the carbonyl
157.
Sadhu, K. M.; Matteson, D. S. Tetrahedron Lett. 1986, 27, 795 and references cited therein.
135
compound. Also, the reaction of (chloromethyl)lithium with the carbonyl compound must
be faster than its decomposition.
We believed that a-halocarbanions
TAS-vinylketenes
such as (chloromethyl)lithium would react with
In fact, in the very
in the desired fashion to afford cyclopentenones.
first reaction tried, (chloromethyl)lithium added to vinylketene 175 to furnish oxsilylcyclopentenone
265 in 65% yield (eq 103).
1.1equiv MeLi-LiBr
1.1 equiv CICH21(293)
THF
(-Pr) 3S 175
-78°C -rt
3h
(I-Pr)3S
265
(103)
65%
265
175
Unfortunately, although several attempts have been made to repeat this reaction, all
efforts have been unsuccessful and resulted in the formation of a complex mixture of
products. Unless otherwise indicated, for all the reactions below (Table 14), 1.0-1.1 equiv
of chloroiodomethane
(293) were used after filtration through alumina, and the
concentration of the reaction mixture was 0.1 M.
Table 14
Entry
Conditions
1
1.1 equiv MeLi-LiBr, THF
THF, -78 °C to rt, 2 h
(293 not filtered)
complex mixture
2
1.0 equiv n-BuLi, THF
-78 °C to rt, 2 h
(293 not filtered)
complex mixture
3
Results
1.2 equiv n-BuLi, THF
complex mixture
-78 °C, 2 h
4
1.0 equiv MeLi-LiBr, THF
-78 °C
136
(distilled 293 and
filtered)
complex mixture
5
1.1 equiv MeLi-LiBr, THF
-78 °C
(new bottle RLi)
complex mixture
6
1.1 equiv MeLi-LiBr, CH 2C12
-78 °C
(different solvent)
complex mixture
7
1.1 equiv MeLi-LiBr, Et20
-78 °C
(different solvent)
complex mixture
1.0 equiv MeLi-LiBr, THF
(concn. = 0.5 M)
complex mixture
8
-78 °C, 2 h
9
1.0 equiv n-BuLi, THF
(lower temperature) complex mixture
-93 °C
10
11
1.0 equiv MeLi-LiBr, THF
-50 °C
(higher temperature)
1.2 equiv MeLi-LiBr, THF
complex
mixture
complex mixture
THF,-78 °C
12
1.2 equiv MeLi-LiBr, THF
(concn. = 0.15 M)
-78 °C, 15 min
13
265<16% yield
complex mixture
1.0 equiv MeLi-LiBr, THF
(added salt)
complex mixture
(added salt)
complex mixture
0.5 equiv LiBF 4 , -78 °C
14
1.0 equiv MeLi-LiBr, THF
0.5 equiv LiBr, -78 °C to rt, 3 h
15
1.0 t-BuLi, THF, -78 °C
complex mixture
Attempts to reproduce the result achieved in the initial reaction included the
following measures:
(1) The reaction was conducted at higher concentrations (entry 8) in
anticipation of trapping the highly unstable x-halocarbanion before it decomposes. (2)
Other bases such as n-BuLi and t-BuLi were used to generate the a-halocarbanion
(entries
2,3,9, and 15). (3) Other solvents were employed (entries 6 and 7). (4) Lithium salts
were added to the reaction mixture to promote the addition to the carbonyl group (entries 13
and 14). (5) Several related
annulation.
158.
-halocarbanions were also examined for the [4 + 1]
However, neither bromomethyllithium
Michnick, T. J.; Matteson, D. S. Synlett 1991, 631.
137
(from CH2Br2/n-BuLi)
158
or
chloromethylmagnesium chloride (from ClCH2I/i-PrMgC) 1
59
produced the desired
cyclopentenone upon reaction with TAS-vinylketene 175 (Table 15).
Table
15
= (i-Pr)3 Si
265
175
i
Fntrv
Cnnditinn-q
1
1.2 equiv n-BuLi, THF
RPCllltC
complex mixture
1.2 CH2Br 2
-78 °C,
1.0 equiv n-BuLi, THF
2
concn = 0.5 M
complex mixture
concn = 0.2 M
complex mixture
1.0 CH 2 Br 2
-78 °C
1.2 equiv n-BuLi, THF
3
-78 C, 1 h
rt, 1.5 h
4
1.0 equiv i-PrMgCl, THF
-78 °C
no reaction
observed by tlc
At this point, the initial result of the [4 + 1] annulation (eq 91) with
chloromethyllithium
has not been reproduced and no explanation as to why this is the case
has been determined.
[4 + 1] Annulation with Sulfur Ylide Reagents
Sulfur ylide reagents such as dimethyloxosulfonium methylide and
dimethylsulfonium methylide are widely used to transfer methylene groups to unsaturated
linkages such as C=O, C=N, C=S, and in some cases C=C.160 We have found that
159.
160.
De Lima, C.; Julia, M.; Verpeaux, J.-N. Synlett 1992, 133.
For reviews on sulfur ylide chemistry, see: (a) Johnson, A. W. In Ylid Chemistry; Academic Press: New
York, 1966.
(b) Trost, B. M.; Melvin, L. S. In Sulfur Ylides; Academic Press: New York,
1975.
(c)
Johnson, C. R. In Comprehensive Organic Chemistry; Jones, D. N., Ed.; Pergamon Press: New York,
1979; pp 247-260. (d) Knipe, A. C. In The Chemistry of the Sulfonium Group; Stirling, C. J. M., Ed.;
John Wiley & Sons: New York, 1981, 313-387. (e) Aube, J. In Comprehensive Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon Press: New York, 1991, pp 819-825.
Sulfur Reagents in Organic Synthesis; Academic Press: London, 1994.
138
(f) Metzner, P.; Thuillier,
A.
dimethylsulfonium methylide 293161is a methylene transfer reagent that reacts smoothly
with TAS-vinylketenes to produce cyclopentenones as shown below.
(i-Pr)3 Si
rC)
+
H2 C-S
293
265
175
In this reaction it is necessary to add the ylide to a solution of the ketene. When the
ylide is present in excess, the desired cyclopentenone product can react further. It is also
noteworthy that when the ylide is present in excess, or when the reaction is conducted at
higher concentrations (> 0.2 M in silylketene), there is evidence of byproduct formation by
tic, and the isolated yield of the desired product is significantly decreased.
Two bases were investigated for the generation of the dimethylsulfonium methylide
from commercially available trimethylsulfonium iodide. Initially, we generated the ylide
from trimethylsulfonium iodide using n-butyllithium in a solution of THF at 0 °C (eq 104).
None of the desired cyclopentenone 265 was formed, and 1 H NMR spectral analysis
suggested that no cyclized product had formed as indicated by the quartet observed at 5.40
ppm for the proton Ha. In contrast, when dimsyl anion (generated from NaH and DMSO)
was used to form the ylide in DMSO at 0 °C (eq 105), the desired cyclopentenone was
produced in 29% yield. In fact, it was found that n-butyllithium could be used to generate
the ylide, if DMSO was used as a polar co-solvent with THF to promote the cyclization.
Under these conditions the [4 + 1] annulation proceeded in 75% yield (eq 106). Lower
yields were obtained when the reaction was conducted in mixtures of THF with other polar
co-solvents such as DMF and DMPU.
161.
(a) Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc. 1965, 87, 1353. (b) Corey, E. J.; Chaykovsky, M.
J. Am. Chem. Soc. 1962, 84, 3782.
139
0
n-BuLi, THF
Me3Sil
(I-Pr)3
O0°C--rt
V
h
(I-Pr)3SI
175
C-O
(i-Pr)3Siy
265
NaH, DMSO
Me 3SI
0°C--rt
N.
1h
(105)
29%
265
175
(I-Pr) 3S1
c/
(104)
1.2 equiv Me2SCH3+1
1.05 equiv n-BuLi
THF-DMSO
0 °C - rt 1 h
O
(I-Pr)3Si
(106)
75%
265
175
One limitation of using DMSO as a co-solvent is that its relatively high melting
point (18.4 °C) prevents the reaction from being conducted at temperatures below 0 °C. In
general, sulfonium ylides have a limited thermal stability, and it is therefore important that it
be possible to conduct the reaction at low temperatures.
It was discovered that DME can
also be used as a solvent, and the annulation can be accomplished in 61% yield (eq 107).
The melting point (-50 °C) of DME is low enough that if necessary this reaction can be
conducted at much lower temperatures than the corresponding THF-DMSO reaction.
(i-Pr)3SI c/
1.2 equiv Me2 SCH3
1.05 equiv n-BuLi
DME
0 C - rt 1.5 h
0
(i-Pr)3Si
(107)
61%
265
175
Dimethylsulfonium methylide can also be generated from trimethylsulfonium iodide
using phase transfer catalysis under aqueous conditions.
162.
Merz, A.; Markl, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 846.
140
16 2
This method of ylide
formation was investigated for reaction with TAS-vinylketene 175. However, as
expected, hydrolysis of the silylketene was a facile process, and this reaction afforded the
carboxylic acid 294 in 24-51% yield depending on the temperature of the reaction mixture.
The acid 294 is characterized by the broad O-H stretching frequency at 3100-2500 cm- 1
and the carbonyl stretching frequency at 1680 cm-1 in the IR spectrum. The structure of
acid 294 was further confirmed by conversion to the methyl ester 295 using TMSdiazomethane (eq 108).163
(I-Pr)3Si C'^°
O
O
1.0 equiv Me3Sil
0.01 equivTBAI
50%NaOH-H2 0, CH2 C12
OH
OCH
OCH3
(108)
920/
24-51%
175
Me3SiCHN2
(i-Pr)3 I
MeOH-benzene
H
rt 2.5 h
294
295
Dimethylsulfonium ylide (generated from trimethylsulfonium iodide and n-BuLi)
also reacted efficiently with other TAS-vinylketenes
(eq 109 and 110). Reactions were
conducted by generating 1.05 equiv of the sulfur ylide in THF and then adding it to a
solution of ketene (0.1 M) in 1:1 THF-DMSO on a scale to afford 60-120 mg of product.
The more stable, less reactive dimethyloxosulfonium
methylidel
61
also added to the TAS-
vinylketene 175 to produce the silylcyclopentenone 265 in 41% yield (not optimized) (eq
1
H1).
163.
Hashimoto, N.; Aoyama, T.; Shiori, T. Chem. Pharm. Bull. 1981, 29, 1475.
141
(i-Pr)3SiCPh
'
1.1equivMe2SCH3+I
1.05 equivn-BuLi
THF-DMSO
0--rt lh
(i-Pr)3Si
(109)
65%
Ph
Ph
182
Et3Si
C/O
296
1.2equivMe2SCH,3+
1.05 equivn-BuLi
THF-DMSO
Et3SI
0-rt 2h
(110)
75%
176
'
1.3equivMe2 S(O)CH3+I
1.4 equivNaH
THF-DMSO
0 C -rt 1 h
43%
297
0
(iPrhS
i
(i'PrSi~
(111)
265
175
We have also recently found that sulfur ylide reagents can serve as effective
"alkylidene transfer" reagents. Specifically, diphenylsulfonium ethylide16 4 was generated
from the commercially available salt, diphenylethylsulfonium fluoroborate (299). This
sulfur ylide added to silylketene 175 in a solution of DME cooled at -50 °C to produce the
desired cyclopentenone
298 in 68% yield (eq 112). The reaction was conducted at low
temperature due to the ylide instability.
Diphenylsulfonium ethylide is known to
decompose in THF with a half life of ca. 5 h at -20 °C and very rapidly at 0 °C. t-BuLi was
found to be the most convenient base for ylide generation. However, it is also possible to
produce the ylide using dichloromethyllithium
(formed via the deprotonation of
dichloromethane with lithium diisopropylamide) at -78 °C. Using this method, the reaction
of the silylketene and diphenylsulfonium ethylide afforded the desired cyclopentenone in
56% yield (eq 112;).
164.
Corey, E. J.; Jautelat, M.; Oppolzer, W. Tetrahedron Lett. 1967, 2325.
142
'
1.1 equiv CH3 CH 2 SPh+BF
2
4 (299)
1.05 equiv BuLi
DME
-50-- -20 °C
h
68%
(i-Pr) Si
3
O
C
1.1 equiv CH3CH2SPh
+BF4
2
1.0 equivLiCHC12
DME
-78 °C- rt 2h
175
(iPr)SiA
.H,b
(112)
298
56%
The structural assignment for a-silylcyclopentenone 298 was again based on the
coupling constant observed between the protons at C-4 and C-5 of the cyclopentenone ring.
As mentioned earlier, the coupling constant for the protons attached at C-4 and C-5 of
trans-4,5-dimethyl-cyclopent-2-en-1-one
(Ha and Hb) is relatively small (Jab = 2.5 Hz).1 5 0
This is similar to the coupling constant observed between Ha and Hb (Jab = 3.2 Hz) for
silylcyclopentenone
298.
Dreiding models of both the cis and trans ax-silyl
cyclopentenones have been constructed. For the cis isomer, it appears that the dihedral
angle between Ha and Hb cannot be much greater than -30 ° , thus a large (Jab = 8-9 Hz)
coupling constant is predicted. In contrast, the dihedral angle between Ha and Hb in the
trans isomer appears to be much larger (-110 ° ) and therefore should result in a much
smaller coupling constant (Jab= 2-3 Hz) which is consistent for the Jab value observed for
cyclopentenone 298.
As mentioned earlier for the reaction of silylvinylketene 175 and TMSdiazomethane (see p. 127), the stereochemical assignment of the C-4 and C-5 substituents
as trans offers insight into the mechanistic pathway of the annulation. It is believed that
initially the sulfur ylide adds to the vinylketene in a stereoselective fashion to form the Z
enolate which can then ionize to produce the pentadienyl cation with the substituents at C-4
and C-5 positioned outward to avoid unfavorable steric interactions as shown below. Ring
closure via a 4n electrocyclic conrotatory process then affords the trans substituted product.
0R3Si
2
143
Ylides in which both R' and R" are alkyl have been noted to be the least stable of
all sulfonium ylides and the most difficult to prepare. There are two possible methods for
the formation of isopropyl ylide 299 (Figure 16).164 The first procedure involves the
alkylation of diphenylsulfonium ethylide (generated from diphenylethylsulfonium
fluoroborate (299) and t-BuLi) with methyl iodide followed by deprotonation using lithium
diisopropylamide.
However, attempts to generate the ylide using this method and then add
it to the silylketene did not lead to the desired cyclopentenone (eq 113).
Figure 16
0
(i-Pr)3S1
C
+ ,Ph
+
/-- BF4
DME,-78 °C
t-BuLi
then Mel
then LDA
o
(i-Pr)3Si
V
e LD
(113)
(113)
Ph
300
175
I
I
265
(50%/)
298
(42%)
Instead, the two products isolated were the methylide addition product 265 (50%)
and the ethylide addition product 298 (42%). Cyclopentenone 298 is presumably
produced via the addition of the unalkylated initial ethylide to the TAS-vinylketene.
Alternatively, decomposition of diphenylethylsulfonium fluoroborate to the methylide can
144
occur via n-elimination as illustrated below (Scheme 28). We believe that the resulting
methylide then adds to ketene 175 to give 265. In this reaction, it is very likely that none
of the desired isopropylide was generated under the reaction conditions.
Scheme 28
lch'
~~~~H
I-CH
Ph
3
,
/Ph
H3C--S
I'
,Ph
H2C=CH2H
Ph
Ph
dimethylsulfonium methylide
An alternative method for the preparation of the isopropylide 299 involves the use
of the crystalline salt, diphenylisopropylsulfonium fluoroborate (301) which can be
formed from diphenyl sulfide, isopropyl iodide, and silver fluoroborate (eq 1 14).165 The
crude salt was recrystallized from dichloromethane-ether two times to give a white
crystalline product in 19% yield (lit. 55%
).165
Ph
AgBF4
PhSPh
+
I
O C -- rt1 2.5 h
19%
S,
BF4
Ph
Ph
(114)
301
Preliminary studies have shown that the reaction of silylvinylketene 182 with
diphenylsulfonium isopropylide (generated via the deprotonation of 301 using tbutyllithium) affords as a viscous oil -silylcyclopentenone 302 in -60-70% yield (eq
115). Unfortunately, the cyclopentenone is inseparable by column chromatography from
other byproducts produced in the reaction and has not yet been isolated in >95% purity.
Studies are presently being conducted in which the annulation product is protodesilylated
(for a discussion of this type of desilylation, see p. 168) prior to purification in attempts to
isolate uncontaminated 303.
165.
Nadeau, R. G.; Hanzlik, R. P. Methods in Enzymology, 1969, 15, 347.
145
(i-Pr)3S1
co
301
tBuLi
TFA
CHC12
(i-Pr)3 S1
THF:DMSO
Ph
-78 -- 0 C
Ph
Ph
Ph
303
302
182
Attempted Synthesis of Heterocycles
In principle, our [4 + 1] annulation could provide efficient access to five-membered
heterocycles in addition to the carbocycles described in the preceding sections. We have
investigated a number of reagents that could potentially function as "nitrenoid" equivalents
to make nitrogen heterocycles. We have focused our attention on a number of organic
azides, which we had hoped would combine with TAS-vinylketenes and afford unsaturated
lactams (eq 116).
I-
O
R3 Si
RN3
,
R
(116)
1
R2
R
I
L
2
R
_I
We first explored the reaction of silylvinylketene 175 with methanesulfonyl azide
(304), an azide reagent that was readily available from our earlier diazo transfer studies.
The ketene and azide 304 were combined in dichloromethane and stirred at room
temperature for 116 h (eq 117). No reaction was observed by tlc analysis, and -80% of
the silylketene was recovered. It is believed that the electron withdrawing sulfonyl group
stabilizes azide 304 to such an extent that it does not readily undergo nucleophilic addition
to the silylketene.
146
(.Pr)
O0
S
3
MsN3 (304)
OC-rt
116h
1
(117)
175
We next turned our attention to an azide in which the nitrogen atom was expected to
be more nucleophilic than MsN 3. Benzyl azide (305) was prepared in 91% yield as shown
below (eq 118).166 When silylketene
175 was combined with benzyl azide in
dichloromethane and stirred for several days at room temperature, no reaction was
observed (eq 119). Catalysis of the reaction using BF3 .OEt2 was also attempted. At -78
°C, no reaction was observed by tlc analysis, but as the reaction warmed to room
temperature, a complex mixture of products formed at -16 °C.
PhCH2CI +
(i-Pr) 3S
C
NaN3
MeOH
65 °C 12 h
91%
PhCH2N3
305
(118)
0
PhCH2N3
Ph
CHC2
(119)
175
175
Commercially available TMS-azide was found to react with silylketene 175 to
produce regioisomeric isocyanates 306 and 307 in 12% overall yield (eq 120). In this
reaction, 33% of the ketene starting material was also recovered, and minor amounts of two
other unidentified products were formed as well. We believe the isocyanates 306 and 307
arise from the pathway depicted in Scheme 29.
166.
Wiley, R. H.; Hussung, K. F. J. Chem. Soc. 1955, 21, 190.
147
(I-Pr)
3S 1c
/0
(CS
Me3SiN3, CH2Cb
0°C- rt 20h
(120)
12%
306
175
(IPr)S
307N307
Scheme 29
Initially, TMS-azide adds to the ketene carbonyl to form enolate 308 which can
undergo protonation with cleavage of the trimethylsilyl group to produce acyl azide 310.
Acyl azides are well known to rearrange to isocyanates, and this process is a key step in a
number of important synthetic methods such as the Curtius and Schmidt rearrangements.
Phenyl azide (313) was prepared from commercially available phenylhydrazine by
the action of nitrous acid upon phenylhydrazine hydrochloride (eq 121).167Thus, phenyl
hydrazine was added to an aqueous HC1 acid solution at -20 °C, and the precipitation of
phenylhydrazine hydrochloride was observed. A small amount of diethyl ether was added
to the reaction mixture, then an aqueous solution of sodium nitrite was added slowly over
time, and the reaction mixture was subjected to steam distillation. Phenyl azide was
extracted from the aqueous phase of the distillate and then distilled under reduced pressure
167.
Lindsay, R. O.; Allen, C. F. H. Organic Syntheses Coil Vol 3, 1955, 710.
148
to yield phenyl azide (313) as a pungent, pale yellow, oily azide in 49% yield (lit. 6568%).167
NHNH2
HCI
NaNO2
49%
+
NaCI
+
2H2 0
(121)
313
Silylketene
175 and phenyl azide (313)
were combined at 0
C in
dichloromethane, and the reaction mixture was allowed to warm to room temperature and
stirred for 48 h (eq 122). No significant reaction was observed between the ketene and
phenyl azide by tic analysis.
0o
(i-Pr)3S1
C 'O
PhN3 (313), CH2C12
(I-Pr)3SI1
n
Ph
(122)
175
Initial attempts to prepare butyl azide (315) from butyl bromide (314) and sodium
azide were unsuccessful (entries 1 and 2; Table 16). In one experiment, 1.0 equiv of butyl
bromide was combined with 2.0 equiv of NaN3 in DMSO at room temperature for 6 h,'6 8
and in methanol at 70 °C for 15 h. 16 9 After filtration of the crude reaction mixture, in both
cases the IR spectrum lacked typical azide stretches (2250-2080 cm- 1). Using phase
transfer catalysis, we were finally able to prepare butyl azide (315) beginning with butyl
bromide and sodium azide using Aliquat 336 (tricaprylylmethylammonium chloride) as the
catalyst (entry 3).170 Butyl azide was isolated after distillation under reduced pressure as a
clear liquid in 39% yield with a characteristic azide IR stretching frequency at 2090 cm- 1.
Table
168.
169.
170.
16
(a) Brown, H. C.; Salunkhe, A. M.; Singaram, B. J. Org. Chem. 1991, 56, 1170. (b) Evans, D. A.;
Weber, A. E. J. Am. Chem. Soc. 1987, 109, 7151.
Lieber, E.; Chao, T. S.; Rao, C. N. R. J. Org. Chem. 1957, 22, 238.
Reeves, W. P.; Bahr, M. L. Synthesis 1976, 823.
149
Br
+
314
Entry
Or
-
e
'
N3
N
'
NaN3
l
315
Conditions
-
_
~-
_
_
.
*
1
DMSO, rt, 6 h
no reaction
2
MeOH, 70 C, 15 h
no reaction
3
Aliquat 336, H2 0
39% yield
rt to 100 °C, 5 h
The addition of butyl azide (315) to silylketene 175 was attempted under a range
of conditions as illustrated below (Table 17). In general, no significant reaction was
observed between the ketene and azide, and -65% of the silylketene was recovered. One
exception included when the temperature was elevated to 174 °C and in this case a complex
mixture of products formed (entry 4).
Table 17
c
0(i-Pr)3Si
(i-PrSix;C
N-(CH2) 3CH 3
(iPr)S
175
Entrv
'"
"~~~
Condition,
-
w_
.
E.
*
1
1.5 equiv BuN 3
C1CH2 CH 2Cl, 60 °C, 5 h
No reaction observed by tic
ketene recovered in 69% yield
2
1.1 equiv BuN 3
benzene, 80 °C, 5 h
No reaction observed by tic
3
1.3 equiv BuN 3
xylene, 140 °C, 2 h
No reaction observed by tic
ketene recovered in 64% yield
1.0 equiv BuN 3
complex mixture observed by tic
4
decane, 174 °C, 2 h
150
Protodesilylation
a-Silylcyclopentenones serve as important intermediates in organic synthesis. We
have investigated the protodesilylation of our [4 + 1] annulation products to establish that
our method can provide access to cyclopentenones lacking a 2-silyl substituent.
Protodesilyation of (trimethylsilyl)vinylsilanes is known to be a relatively facile process. 171
However, difficulty can arise in the removal of more bulky silyl substituents such as the
triisopropylsilyl group. In fact, optimization of the desilylation of our triisopropylsilylsubstituted cyclopentenones required considerable experimentation. A number of acidic
and fluoride-based reagents were examined (Table 18).
Table 18
O
O
265
316
(I-Pr) 3 Si
171.
Entry
Conditions
Results
1
20 equiv TFA, CH 2C12, rt, 2 h
then MeOH, rt, 3 h
no reaction observed by tic
2
20 equiv TFA, CH 3 CN, rt, 2 h
then MeOH, rt, 3 h
no reaction observed by tic
3
20 equiv TFA, CDC13 , rt, 2 h
MeOH, rt, 3 h
no reaction observed by tic
4
HF py, THF, rt
no reaction observed by tic
5
48% HF, CH 3 CN, rt, 22 h
no reaction observed by tic
For a review on protodesilylation of vinylsilanes, see: Fleming, I.; Dunogues, J.; Smithers, R. Organic
Reactions, 1989, 37, 93.
151
6
HBF 4 , CH 3 CN, rt, 22 h
no reaction observed by tic
7
HCI-CHC1 3 , rt, 19 h
no reaction observed by tlc
8
BF 3 -2AcOH, CH 2 C12
0 °C to rt, 9 h
no reaction observed by tic
9
20 equiv MsOH, MeOH
rt, 6 h
complete conversion by tic
to more polar product
In contrast to the other procedures listed, protodesilylation using methanesulfonic
acid was found to be an efficient process even in the case of the bulky triisopropylsilyl
group. For example, desilylation of 265 proceeds in 95% yield on exposure to 16 equiv
of methanesulfonic acid in methanol at room temperature to afford cyclopentenone 316172
(eq 123). Efforts to reduce the number of equivalents of acid to <10 equiv were
unsuccessful, and the complete conversion of starting material to product was not observed
by tic. Likewise, the triethylsilyl substituted cyclopentenone undergoes desilylation with
10 equiv of methanesulfonic acid in 89% yield (eq 124).
0
16 equiv MsOH
MeOH
rt 3 h
95%
(i-Pr)
(123)
316
265
0
10 equiv MsOH
MeOH
Et3Si
rt 8.5 h
89%
b
316
(124)
297
Summary
In summary, we have presented a new strategy which provides an efficient and
convenient route to a variety of substituted TAS-vinylketenes based on the photochemical
Wolff rearrangement. We have found these substituted vinylketenes to be remarkably
robust substances and have investigated their reactivity as four-carbon building blocks in
172.
Conia, J.-M.; Leriverend, M.-L. Bull. Soc. Chim. Fr. 1970, 2981.
152
synthesis. TAS-vinylketenes serve as reactive dienes in the Diels-Alder reaction with a
number of reactive dienophiles. We have found that TAS-vinylketenes also participate in
novel [4 + 1] cycloadditions with a number of "carbenoid" reagents which provide an
interesting new route to 5-membered carbocyclic products. Overall, we anticipate that
improved accesss to substituted TAS-vinylketenes via the photo Wolff strategy will
enhance the synthetic utility of silylvinylketenes, and we will continue to explore their
promising new applications in synthesis.
153
154
Part
III
Experimental Section
155
General Procedures
All reactions were performed in flame-dried glassware under a positive pressure
of argon or nitrogen.
Reaction mixtures were stirred magnetically unless otherwise
indicated, except for photochemical reactions which were not stirred unless otherwise
indicated. Air and/or moisture sensitive reagents and solutions were transferred via syringe
or cannula and were introduced into reaction vessels through rubber septa. Reaction
product solutions were concentrated with the use of a Biichi rotary evaporator at
approximately 20 mm Hg unless otherwise indicated.
Materials
Commercial grade reagents and solvents were used without further purification
except as indicated below.
Distilled under nitrogen, argon, or vacuum from calcium hydride: acetonitrile,
dichloromethane,
diisopropylamine,
diisopropylethylamine,
dimethyl
acetylenedicarboxylate, 1,1,1,3,3,3-hexamethyldisilazane, 2,2,6,6-tetramethylpiperidine,
dibromomethane, dichloroethane, triethylamine, triisopropylsilyl chloride, trisopropyl-,
tert-butyldimethyl-, and triethylsilyl trifluoromethanesulfonate.
Distilled under nitrogen, argon, or vacuum from sodium benzophenone ketyl or
dianion: benzene, diethyl ether, dimethyl sulfoxide, ethylene glycol dimethyl ether (DME),
tetrahydrofuran, and toluene
156
Distilled under argon or nitrogen: 1-acetyl-1-cyclohexene, 4-(2,6,6-trimethyl1-cyclohexen-1-yl)-3-buten-2-one, 4-(2,6,6-trimethyl- 1-cyclohexen-2-yl)-3-buten-2-one,
and trans-3-penten-2-one.
Purification of other reagents was accomplished in the following manner;
methyl iodide was distilled under argon and filtered through alumina.
Alkyllithium reagents were titrated in tetrahydrofuran or hexane at 0 °C with
sec-butanol using 1,10-phenanthroline as an indidcator. 17 3
Chromatography
Analytical thin-layer chromatography (TLC) was performed on Merck precoated glass-backed silica gel 60 F-254 0.25 mm plates. Visualization of spots was
effected by one or more of the following techniques: (a) ultraviolet irradiation, (b)
exposure to iodine vapor, (c) immersion of the plate in a 10% solution of phosphomolybdic
acid in ethanol followed by heating to ca. 200 °C, (d) immersion of the plate in an ethanolic
solution of 3% p-anisaldehyde containing 0.5% concentrated sulfuric acid followed by
heating to ca. 200 °C, and (e) immersion of the plate in an ethanolic solution of 3% pvanillin containing 0.5% concentrated sulfuric acid followed by heating to ca. 200 °C.
Column chromatography was performed by using 230-400 mesh Merck or
Baker silica gel.
173.
Watson, S. C.; Eastham, J. F. J. Organomet. Chem. 1967, 9, 165.
157
Instrumentation
Photolyses were performed in a Rayonet Photochemical Reactor Model RPR100 or RPR-200, both produced by the Southern New England Ultraviolet Company. The
reactors each contained sixteen low pressure mercury bulbs of 253.7 or 300 nm ultraviolet
light. The photolyses were conducted with an internal fan in operation, and the internal
temperature of the chamber was never higher than 35 °C.
Melting points were determined with a Fisher-Johns melting point apparatus
and are uncorrected. Boiling points are uncorrected.
Infrared spectra (IR) were recorded using a Perkin-Elmer 1320 grating
spectrophotometer.
1H
NMR specta were recorded with a Varian XL-300 (300 MHz) and a Varian
Unity 300 (300 MHz) spectophotometer.
Chemical shifts are expressed in parts per million
(6) downfield from tetramethylsilane.
13 C
NMR spectra were determined on a Varian XL-300 (75 MHz)
spectrophotometer.
Chemical shifts are reported in parts per million (), relative to
tetramethylsilane (with the central peak of CDC13 at 77.0 ppm used as a standard).
Ultraviolet-visible spectra were recorded with a Perkin-Elmer 552 UV-vis
spectrophotometer, and absorbances are reported in nanometers (nm).
High resolution mass spectra (HRMS) were measured on a Finnegan MATT8200 spectometer.
Elemental analyses were performed by Robertson Microlit Laboratories, Inc.,
of Madison, New Jersey.
158
159
CO2 H
Br
CH 3
35
1-(5-Methyl-l-naphthoyl)-ethanone
36
(36).
A 500-mL, three-necked, round-bottomed flask equipped with an argon inlet
adapter, rubber septum, and a mechanical stirrer was charged with a solution of 5-bromol-napthoic acid 35 (6.09 g, 24.2 mmol) in 200 mL of tetrahydrofuran and cooled at 0 °C
while methyllithium solution (1.53 M in diethyl ether, 16.0 mL, 24.5 mmol) was added
dropwise over 5 min. The resultant carboxylate salt solution was stirred at 0 °C for 5 min
and then cooled at -78 C while tert-butyllithium solution (1.70 M in pentane, 29.5 mL,
50.2 mmol) was added dropwise via syringe over 5 min. The green suspension was
stirred at -78 °C for 2.5 h, methyl iodide (4.07 g, 1.80 mL, 28.7 mmol) was added in one
portion, and the orange reaction mixture was allowed to warm to 25 °C over a 2 h period.
The solution was next cooled at -78 °C while methyllithium solution (1.53 M in diethyl
ether, 31.2 mL, 47.8 mmol) was added over 5 min, and the reaction mixture was stirred
for 20 h at 25 °C. The brown reaction mixture was cooled to 0 °C and freshly distilled
chlorotrimethylsilane
(51.9 g, 60.7 mL, 478 mmol) was rapidly added over ca. 1 min. The
ice bath was then removed and the reaction mixture was allowed to come to room
temperature and then transferred via cannula into 300 mL of a rapidly stirring 10% HC1
solution. The organic layer was separated and the aqueous phase was extracted with three
150 mL portions of diethyl ether. The combined organic phases were washed with 200 mL
of' 10% HC1solution, 200 mL of water, and 200 mL of saturated NaCl solution, dried over
M gSO4, filtered, and concentrated to afford 4.96 g of a brown oil.
160
Column
chromatography on 100 g of silica gel (gradient elution with 0-5% EtOAc-hexane) provided
3.73 g (83%) of the methyl ketone 36 as a yellow oil.
IR (CC14):
3060, 2990, 2960, 2880, 1950, 1690, 1605, 1590,
1520, 1475, 1410, 1390, 1360, 1270, 1240, 1210,
1180, 1170, 1090, 1040, 1020, 1000, 970, and 800
1H
NMR (300 MHz, CDC13 ):
8.51 (d, J=8.5 Hz, 1H), 8.13 (d, J=8.4 Hz, 1H), 7.83
(d, J=8.3 Hz, 1H), 7.48 (t, J=8.4 Hz, 1H), 7.45 (t, J=8.4
Hz, 1H), 7.33 (d, J=7.1 Hz, 1H), 2.70 (s, 3H), and 2.66
s, 3H)
13C NMR (75 MHz, CDC13 ):
202.7, 136.6, 134.6, 133.2, 130.2, 128.7, 127.7, 127.6,
127.3, 124.2, 30.0, and 19.5 (one carbon overlapping with
another peak)
UV max (CC14)
287 ( = 2 300) nm
Elemental Analysis:
Calcd for C 1 3 H1 2 0:
Found:
161
C, 84.75; H, 6.57
C, 84.48; H, 6.59
0
IZ
.0
o
o
-cu
Oo
0
i
i
I m--7
o0
-
toCO)
0
- oLO
I
0n
-w 0
-0
_
L
i~~~~
11
to0
-O 0
- o
or
I
o
IUL
I IL
m
N2
0
C
CH 3
36
2-Diazo-1-(5-methyl-1-naphthoyl)-ethanone
0
30
(30).
A 500-mL, three-necked, round-bottomed flask equipped with an argon inlet
adapter, a glass stopper, and a magnetic stir bar was charged with a solution of
1,1,1,3,3,3-hexamethyldisilazane
(2.60 g, 3.40 mL, 16.1 mmol) in 75 mL of
tetrahydrofuran and cooled at 0 °C while n-butyllithium solution (2.70 M in hexane, 5.40
mL, 14.7 mmol) was added dropwise over 3 min. The resultant solution of LiHMDS was
stirred at 0 °C for 10 min, then cooled at -78 C while a solution of 1-(5-methyl-1naphthoyl)-ethanone
36 (2.47 g, 13.4 mmol) in 25 mL of tetrahydrofuran was transferred
dropwise via cannula over ca. 5 min. The enolate solution was stirred at -78 °C for 30 min
and then 2,2,2-trifluoroethyl trifluoroacetate (3.22 g, 2.20 mL, 16.1 mmol) was added
rapidly in one portion. The reaction mixture was stirred at -78 "C for 10 min and then
partitioned between 100 mL of 5% HCl solution and 100 mL of diethyl ether. The organic
phase was separated and washed with 100 mL of saturated NaCl, dried over Na2SO4,
filtered, and concentrated to afford a yellow oil. The oil was dissolved immediately in 45
mL,of acetonitrile and a rubber septum was fitted to the flask. Water (0.240 g, 0.240 mL,
13.4 mmol), triethylamine (2.03 g, 2.80 mL, 20.1 mmol), and methanesulfonyl azide
(2.43 g, 2.30 mL, 20.1 mmol) were added and the resultant yellow solution was stirred at
25 °C for 7.5 h. The solution was poured into a separatory funnel containing 100 mL of
diethyl ether. The organic layer was separated and washed with three 75 mL portions of
10% NaOH solution, 75 mL of water, 75 mL of saturated NaCl solution, dried over
163
Na2SO4, filtered, and concentrated to afford 2.60 g of a yellow solid.
Column
chromatography on 20 g of silica gel (gradient elution with 0-10% EtOAc-hexane) provided
2.36 g (84%) of the
-diazo ketone 30 as a bright yellow solid: mp 84-85 °C.
IR (CC14):
3040, 2950, 2100, 1615, 1600, 1540, 1510, 1460,
1330, 1230, 1155, and 1185 cm - 1
1H
8.30 (br m, 1H), 8.11 (d, J=8.4 Hz, 1H), 7.58 (d,
J=7.1, 1H), 7.47 (t, J=7.8 Hz, 1H), 7.44 (t, J=7.8
NMR (300 MHz, CDCl 3 ):
Hz, 1H), 7.35 (d, J=6.9 Hz, 1H), 5.68 (s, 1H), and
2.69 (s, 3H)
1 3C
NMR (75 MHz, CDC13):
UV max (CC14):
190.0, 136.3, 134.5, 133.0, 129.9, 127.7, 127.3,
127.1, 125.3, 124.3, 123.6, 57.2, and 19.7
260 (e = 10 500), and 303 (8 100) nm
164
I
- a
-C
-c'
,-----=i
-M
o
a
CO
-LO
0
U~~~~~~~~~~~~~~~
'I--
L
InD
co
Cl
165
---
O
,ra
N2
iPr3SiO
0
OSi t-BuMe2
bPr3SiO-C C -COSi
t-BuMe
2
OH
31
CH 3
30
CH 3
29
4- Hydroxy-3- ((S)- 1-tert-butyldimethylsilyloxy-2-propyl)-7-methyl-2-
triisopropylsilyloxy-phenanthrene
(29).
A 30 cm vycor tube ( 9 mm O.D., 7 mm I.D.) was charged with a solution of the
a-diazo ketone 30 (0.147 g, 0.699 mmol) and silyloxyacetylene 31 (0.365 g, 0.985
mmol) in 2 mL of 1,2-dichloroethane. The tube was fitted with a rubber septum and a
second rubber septum (inverted) was secured with wire to the tube to insure a good seal.
The reaction mixture was degassed (three freeze-thaw cycles at -196 °C, < 0.5 mm Hg) and
then irradiated with 253.7 nm light for 8 h in a Rayonet reactor. The resulting solution was
diluted with 1 mL of additional 1,2-dichloroethane and then heated for 8-12 h in a 90 °C oil
bath. Concentration afforded 0.58 g of an orange-brown oil. Column chromatography on
15 g of silica gel (gradient elution with 0-1% EtOAc-hexane) provided 0.27 g (70%) of 29
as a yellow oil []D 28.30 (c = 0.12, CHC13 ).
IR (thin film):
3200, 2960, 2890, 1930, 1825, 1725, 1635, 1610,
1590, 1530, 1505, 1475, 1380, 1335, 1300, and
1285 cm - 1
1H NMR (300 MHz, CDC13):
9.89, (s, 1H), 8.28 (d, J=8.5 Hz, 1H), 8.24 (d,
J=9.4 Hz, 1H), 7.74 (d, J=9.4 Hz, 1H), 7.59 (s,
1H), 7.43 (t, J=7.8 Hz, 1H), 7.38 (d, J=7.0 Hz,
1H), 3.92-4.06 (m, 3H), 2.73 (s, 3H), 1.40-1.48
(m, 3H), 1.17 (d, J=7.3 Hz, 18 H), 1.00 (s, 9H),
0.21 (s, 3H), and 0.17 (s, 3H)
13C NMR (75 Mz, CDC13):
153.6, 153.2, 134.8, 131.6, 131.0, 129.5, 127.4,
125.4, 121.5, 121.0, 119.0, 118.9, 118.8, 101.8,
166
69.0, 31.6, 25.6, 19.7, 18.2, 18.0, 14.9, 12.9,
-5.9, and 6.0
UV max (CC14):
266 (e = 15 800) and 311 (5 000) nm
HRMS:
For C33 H 520 3 Si2 :
167
Calcd: 552.3455
Found: 552.3432
xEL
a.
0
I
0
i~~~~i~
I
0o
Cu
N
_0
m
0
0
In
0
'O
I
0
_
N
I
L-
-
-o
-
-
\
-o
- )
L
-o
0
168
I-Pr 3 SiO
HO
I
N. N.
C'a
OH
OH
0
OSIt-BuMe 2
_
CH3
29
5
3-Hydroxy-2-((R)-l-hydroxy-2-propyl)-8-methyl-1,4-phenanthren-quinone
(5, (+) danshexinkin
A).
A 250-mL, round-bottomed flask fitted with a rubber septum was charged with a
solution of the phenol 29 (0.890 g, 1.61 mmol) in 50 mL of dichloromethane and cooled at
0 "C while a stream of oxygen was bubbled through the mixture via a syringe needle and
tetra-n-butylammonium fluoride solution (1.0 M in tetrahydrofuran, 3.20 mL, 3.20 mmol)
was added dropwise over 2 min. The cooling bath was removed and the reaction mixture
was allowed to warm to 25 °C and stir under a stream of oxygen for 18 h. The reaction
mixture was partitioned between 150 mL of 5% HCl solution and 150 mL of
dichloromethane and the aqueous phase was separated and extracted with two 100 mL
portions of dichloromethane. The combined organic phases were washed with 100 mL of
5% HCl solution, 100 mL of water, 100 mL of saturated NaCl solution, dried over
MgSO4, filtered, and concentrated to afford 1.43 g of an orange solid. Low temperature (-
78 °C) recrystallization from diethyl ether afforded 0.22 g (47%) of (+)-danshexinkun A
(5) as orange crystals: mp 200-201 C (lit 200 °C), 9a [a]D =30.0 0 (c =0.01, CHC13 )
IR (CC14):
3340, 3040, 2980, 1660, 1630, 1595, 1580, 1470,
1400, 1375, 1320, and 1220 cm- 1
1H
9.43 (d, J=8.8 Hz, 1H), 8.43 (d, J=8.8 Hz, 1H),
8.27 (d, J=8.8 Hz, 1H), 8.05 (br s, 1H), 7.64 (dd,
J=7.0, 8.7 Hz, 1H), 7.48 (d, J=7.0 Hz, 1H), 4.01
(dd, J=7.8, 10.9 Hz, 1H), 3.87 (dd, J=5.3, 10.9
Hz, 1H), 3.50-3.52 (m, 1H), 2.74 (s, 3H), and
1.34 (d, J=2.4 Hz, 3H)
NMR (300 MHz, CDC13):
169
13C NMR (75 MHz, DMSO-d 6 ):
185.2, 184.3, 156.8, 135.5, 134.7, 133.0, 131.9,
130.2, 129.7, 129.4, 125.1, 124.8, 122.2, 122.0,
63.7, 32.7, 19.3 and 14.6
UV max (MeOH):
288 ( = 24 000), 330 (6 900), and 375 (4 000) nm
Elemental Analysis:
Calcd for C1 8 H 1 6 04: C, 72.96; H, 5.44
Found:
170
C, 73.25; H, 5.37
S
a.
-a
~0o
o
so
o
00
0
C',
CD
0
o
0
- oo
UN
_
o
U,
V)
-O
_o
D
o
No
0
_I
q
171
T
0
0O
HO
0
0
CH3
OH
0
CH3
5
9
3,4-Dihydro-3-(R)-methyl-[3,2-c]-furo-8-methylphenanthra-1,2-dione
(9,
dihydrotanshinone-I).
A 100-mL, round-bottomed flask was charged with a heterogeneous solution of
danshexinkun A (0.921 g, 3.11 mmol) in 40 mL of ethanol. To this solution was added 12
ml of concentrated H2 SO4 over 5 min (CAUTION! EXOTHERMIC REACTION!) and
the resultant deep red mixture was stirred at 25 C for 30 min. The reaction mixture was
poured into 200 mL of water and extracted with three 200 mL portions of diethyl ether.
The combined organic phases were washed with two 200 mL portions of 10% HC1
solution, 100 mL of saturated NaHCO3 solution, dried over MgSO4, filtered, and
concentrated to afford 0.79 g of a red solid. Column chromatography on 30 g of silica gel
(elution with CHC13 ) afforded 0.55 g (65%) of (-)-dihydrotanshinone
mp 221-222
C (lit 224-225
C);
1
I (9) as a red solid:
[a]D -300 ° (c = 0.01, CHC13,) (lit. [a]D -328 ° (c =
0.11, CHC1 3)) 1 1
IR (CC14):
3050, 2990, 1655, 1630, 1590, 1540, 1510, 1470,
1420, 1400, 1370, 1350, 1330, 1300, 1260, 1190,
1170, 1150, 1020, 995, 955, 930, and 895 cm- 1
1H
9.32 (d, J=8.8 Hz, 1H), 8.33 (d, J=8.9 Hz, 1H),
NMR (300 MHz, CDC13):
7.79 (d, J=8.4 Hz, 1H), 7.59 (dd, J=7.0, 8.8 Hz,
1H), 7.42 (d, J=7.1 Hz, 1H), 4.98 (app t, J=9.6
Hz, 1H), 4.44 (dd, J=5.9, 9.4 Hz, 1H), 3.61 (ddq,
J=6.6, 9.9, 7.0 Hz, 1H), 2.72 (s, 3H), and 1.42 (d,
J=7.1 Hz, 3H)
172
13C NMR (75 MHz, CDC13 ):
184.3, 175.7, 170.6, 134.9, 134.8, 132.0, 130.4,
128.8, 128.2, 126.1, 125.0, 120.3, 118.4, 81.6,
53.4, 34.7, 19.9, and 18.8
IUV max (MeOH)::
218 (c = 19 500), 262 (28 800), 270 (22 900), 290
(7 100), and 353 (2 800) nm
Elemental Analysis:
Calcd for C 18 H 14 0 3 : C, 77.00; H, 6.80
C, 77.21; H, 6.78
Found:
173
xa-
Lo
to0
a.
c
0
.n
0
C
v
*m
0
o
-
air0
0o
o
cn
C
-N
C
-w
C
0*
174
0
CH3
CH3
0
0
CH3
0
CH3
9
3,8-Dimethyl-[3,2-c]-furophenanthra-1,2-dione
0
1
(1, tanshinone-I).
A 100-mL, three-necked, round-bottomed flask equipped with an argon inlet, glass
stopper and rubber septum was charged with a solution of dihydrotanshinone-I (9, 0.426
g, 1.53 mmol) and 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ, 0.700 g, 3.08
mmol) in 40 mL of benzene. The reaction mixture was stirred at 25 C for 75 h. An
additional portion of DDQ (0.073 g, 0.319 mmol) was added and the reaction mixture was
stirred for 20 h, filtered, and concentrated. Column chromatography on 12 g of silica gel
(elution with 30% CHC13-benzene) afforded 0.21 g (48%) of tanshinone-I (1) as dark red
needles: mp 226-227 °C (lit 233-234 °C)."
IR (KBr):
3100, 2950, 2900, 1650, 1580, 1535, 1490, 1420,
1390, 1370, 1320, 1260, 1170, and 1150 cm-l1
1H
9.22 (d, J=8.5 Hz, 1H), 8.27 (d, J=8.8 Hz, 1H),
NMR (300 MHz, CDC13):
7.78 (d, J=8.9 Hz, 1H), 7.52 (dd, J=6.9, 8.8 Hz,
1H), 7.32 (d, J=7.0 Hz, 1H), 7.22 (s, 1H), 2.66 (s,
3H), and 2.25 (s, 3H)
13C NMR (75 MHz, CDC13):
183.4, 175.5, 161.1, 142.0, 135.2, 133.6, 132.8,
130.6, 129.5, 128.3, 124.7, 123.0, 122.0, 121.7,
120.4, 118.7, 19.8, and 8.8
UV max (CC14):
266 ( = 15 800) and 311 (5 000) nm
HRMS:
For C 3 3H 5203Si 2 :
175
Calcd: 552.3455
Found: 552.3452
-Q.
X
o
o
o
-o
-0.
cu
C
m
C
C
Ao
in
c
O
(0
_N
_,C
o
_o
- d
c
C
176
0
HO
t-BuMe 2 SIO
OMe
OMe
(S)-(+)-Methyl-3-tert-butyldimethylsilyloxy-2-methylpropionate
(39).
A 100-mL, three-necked, round-bottomed flask equipped with an argon inlet
adapter, a glass stopper and a rubber septum was charged with a solution of (S)-(+)methyl-3-hydroxy-2-methylpropionate
36 (2.57 g, 2.40 mL, 21.5 mmol) in 50 mL of
dichloroethane and cooled at 0 °C in an ice bath while imidazole (total: 3.50 g, 51.6 mmol)
and tert-butyldimethylsilyl
chloride (total: 3.90 g, 25.8 mmol) were added alternately in
small portions over ca. 15 min. The reaction mixture was stirred for 12 h allowing the ice
bath to warm to room temperature, and poured into 60 mL of saturated NaC1 solution and
extracted with two 100 mL portions of diethyl ether. The combined organic phases were
washed with three 100 mL portions of water and one time with 100 mL of saturated NaCl
solution, dried over Na2SO4, filtered, and concentrated to afford a clear liquid. Filtration
through silica gel (elution with 5% EtOAc-hexane) and concentration afforded 5.03 g
(100%) of the silyloxyester 39 as a colorless liquid.30C []D=+15.5
IR (thin film):
1H NMR
13 C
(300 MHz, CDC13):
NMR (75 MHz, CDC13):
(c=1.25, CHC13 )
2975, 2945, 2880, 1750, 1470, 1440, 1400, 1370,
1265, 1200, 1180, 1100, 1030, 1010, 995, 950,
840, and 790 cm-l1
3.73 (dd, J=6.4, 10.1 Hz, 1H), 3.62 (s, 3H), 3.573.62 (m, 1H), 2.60 (app q, J=7.0 Hz, 1H), 1.08 (d,
J=6.1 Hz, 3H), 0.82 (s, 9H), and 0.01 (s, 6H)
175.7, 65.2, 51.3, 42.4, 25.5, 17.9, 13.2, and -5.9
177
I
c
-
O
Co
iI
c
i'-
--c
C
C
Cru
ii
C
_
tT
C-~~~~~~~~~~~~~~~~~
C
c
20o
.L
Co
-cC
_
Lo
o
_C
m
_CC
_
i
I--
i
C
i
_(
I
I
C
-0
_C
-
178
)
O
t-BuMe
2 SiO
~OS('Pr)3
_
t-BuMe
2SiO
OMe
39
SI
31
2-((R)-1-tert-B utyldimethylsilyloxy-2-propyl)- 1-triisopropylsilyloxyacetylene
(31).
A 500-mL, three-necked, round-bottomed flask equipped with an argon inlet
adapter, a rubber septum, and a glass stopper was charged with a solution of
dibromomethane (5.70 g, 2.30 mL, 33.1 mmol) in 40 mL of tetrahydrofuran and cooled at
-78 °C in a dry ice-acetone bath.
A 100-mL, three-necked round-bottomed flask equipped with an argon inlet
adapter, a rubber septum and a glass stopper was charged with a solution of 2,2,6,6tetramethylpiperidine (5.36 g, 6.40 mL, 37.7 mmol) in 40 mL of tetrahydrofuran and
cooled at 0 °C, while n-butyllithium solution (2.70 M in hexane, 13.2 mL, 35.6 mmol) was
added dropwise over ca. 5 min. The LiTMP solution was stirred at 0 °C for 15 min, then
cooled at -78 C and transferred dropwise via cannula into the bromide solution over ca. 10
min and the resulting yellow solution was stirred at -78 °C for 15 min. A precooled (-78
°C) solution of the silylester 39 (3.52 g, 15.1 mmol) in 40 mL of tetrahydrofuran was
added dropwise via cannula over 15 min, and the solution was then stirred for 15 min at
-78 °C. n-Butyllithium solution (2.70 M in hexane, 27.3 mL, 73.8 mmol) was added over
5 min, the cooling bath was removed, and the reaction mixture was allowed to stir at 25 °C
for 45 min. The solution was cooled at -78 °C while a solution of triisopropylsilyl chloride
(14.8 g, 16.4 mL, 76.8 mmol) in 40 mL of tetrahydrofuran was added via cannula over ca.
10 min. The dry ice-acetone bath was replaced with an ice bath and the reaction mixture
was stirred at 0 °C for 5 h. The reaction mixture was partitioned between 100 mL of
pentane and 100 mL of saturated NaHCO3 solution, and the aqueous layer was separated
and extracted with 100 mL of pentane. The combined organic phases were washed with
179
200 mL of water, 200 mL of saturated NaCI solution, dried over Na2SO4, filtered, and
concentrated initially at 20 mm Hg, then at <0.005 mm Hg for 48 h to afford an orange oil.
The crude product was divided into three equal portions and column chromatography on
silica gel (elution with hexane) afforded 1.09 g (20%) of the silyloxyacetylene
31 as a
colorless oil.
IR (thin film):
2960, 2880, 2990, 1460, 1385, 1365, 1305, and
1250 cm-1
1H
3.62 (dd, J=5.2, 10.0 Hz, 1H), 3.31 (app t, J=9.0
Hz, 1H), 2.40-2.52 (m, 1H), 1.18-1.28 (m, 3H),
1.11 (d, J=6.7 Hz, 18 H), 1.05 (d, J=10.3 Hz, 3H),
0.89 (s, 9H), and 0.04 (s, 6H)
NMR (300 MHz, CDC13):
1 3C
NMR (75 MHz, CDC13 ):
88.0, 68.4, 32.5, 27.7, 25.7, 18.5, 18.1, 17.1,
11.6, and -5.7
HRMS:
Calcd for C16 H33 02Si2 (M+-C4 H9): 313.2017
Found:
313.2019
180
i
-a
aC.
0o
i
I
-I
o
7=:D
i
cN
O
A
C
m
C-
v
CL
U
-
c
0
-0
o
C
_C
0(5
0
181
0
0
133
131
General Procedure for Diazo Transfer.
ethanone
1-Diazo-2-(1-cyclohexenyl)-2-
(133).
A 50-mL, three-necked, round-bottomed flask equipped with a rubber septum,
argon inlet adapter, and a glass stopper was charged with a solution of 1,1,1,3,3,3,hexamethyldisilazane (1.0 mL, 0.781 g, 4.84 mmol) in 10 ml of THF and then cooled at 0
°C in an ice-water bath while n-butyllithium solution (2.70 M in hexane, 1.6 mL, 4.43
mmol) was added rapidly dropwise over 2 min. After 10 min, the resulting solution was
cooled at -78 C in a dry ice-acetone bath while a solution of 1-acetyl-1-cyclohexene (0.52
mL, 0.500 g, 4.03 mmol) in 8 mL of THF was added dropwise over 5 min (the flask was
rinsed with 1 mL of additional THF). The reaction mixture was stirred at -78 C for 30
min, and then 2,2,2-trifluoroethyl trifluoroacetate (0.65 mL, 0.949 g, 4.84 mmol) was
added rapidly (1 s) by syringe in one portion. After 10 min, the reaction was poured into a
separatory funnel containing 20 mL of 5% aqueous HCl solution and 25 mL of Et2O. The
aqueous phase was extracted with two 15 ml portions of Et2O, and the combined organic
phases were then washed with 20 mL of saturated NaCl solution, dried over Na2SO4, and
concentrated at reduced pressure to give 1.90 g of a yellow oil which was immediately
dissolved in 15 mL of CH3 CN and transferred to a 50-mL, three-necked, round-bottomed
flask equipped with a rubber septum, argon inlet adapter and a glass stopper. Water,
(0.073 mL, 0.073 g, 4.03 mmol), Et3 N (0.84 mL, 0.612 g, 6.05 mmol), and
methanesulfonyl azide (0.70 mL, 0.733 g, 6.05 mmol) were added rapidly dropwise in that
order. The resulting solution was stirred at room temperature for 3 h and then diluted with
50 mL of Et2O and washed with three 15-mL portions of 10% aqueous NaOH solution,
182
three 15-mL portions of H20, and 20 mL of saturated NaCI solution, dried over MgSO4,
filtered, and concentrated to afford 0.51 g of a yellow oil. Column chromatography on 10
g of silica gel (elution with 10% EtOAc-hexane) provided 0.48 g (80 %) of the diazo
ketone 133 as a yellow solid, mp 28-28.5 °C, with spectral characteristics identical with
those reported previously.2 9
183
o
0
N2
.
SI(/-Pr)3
133
160
i-Diazo-2-(1-cyclohexenyl)-l-triisopropylsilyl-2-ethenone
(160).
A 50-mL, three-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 133 (0.403 g,
2.69 mmol) in 20 mL of a 50:50 solution of Et2O-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (0.47 mL, 0.349 g, 2.70 mmol) was added dropwise over 1
min. Triisopropylsilyl trifluoromethanesulfonate (0.72 mL, 0.820 g, 2.69 mmol) was then
added dropwise over 1 min and the resulting solution was stirred for 4 h while the ice-
water bath warmed to room temperature. The reaction mixture was filtered through Celite
with the aid of 5 mL of Et2 O and the filtrate was concentrated to afford a yellow oil.
Column chromatography on 10 g of silica gel (gradient elution with 0-2% EtOAc-hexane)
provided 0.62 g (75%) of the silyldiazo ketone 160 as a yellow oil.
IR (film):
2930, 2870, 2240, 2060, 1595, 1460, and 1365
cm-1
1H
13
NMR (300 MHz, CDC13):
C NMR (75 MHz, CDC1 3):
6.32 (m, 1H), 2.25-2.26 (m, 2H), 2.18-2.20 (m,
2H), 1.64-1.67 (m, 4H), 1.35 (sept, J = 7.1 Hz,
3H), and 1.11 (d, J = 7.3 Hz, 18H)
196.0, 138.0, 132.5, 49.7, 25.1, 24.7, 22.0, 21.7,
18.5, and 11.5
12V max (hexane):
292 ( = 2700) and 220 (8400) nm
HRMS:
Calcd for C17 H3 0 ON2 Si:
Found:
184
306.2127
306.2129
X
.
.0
-o
i
-
Ir
-
mmmmm
-m
K
-.El'
I
-o
z
0
Z&
T-
-t
-Co
185
-0
o
(-Pr) 3
N2
Pr) 3
174
160
2-(1-Cyclohexenyl)-2-triisopropylsilylketene
(174).
A solution of silyldiazo ketone 160 (1.16 g, 3.78 mmol) in 38 mL of benzene was
distributed evenly between two 25-cm vycor tubes (15 mm O.D., 13 mm I.D.) fitted with
nibber septa. A second rubber septum (inverted) was secured with wire to each tube to
insure a good seal, and each solution was degassed (three freeze-pump-thaw cycles at -196
°C, < 0.5 mm Hg) and then irradiated with 300 nm light for 4 h in a Rayonet reactor. The
resulting solutions were combined and concentrated to afford 2.20 g of an orange-brown
oil. Column chromatography on 15 g of silica gel (elution with hexane) provided 0.94 g
(89%) of ketene 174 as a viscous yellow oil.
IR (thin film):
2936, 2866, 2076, 1676, 1641, and 1461 cm-l1
1H
5.40 (t, J = 3.0 Hz, 1H), 2.09-2.12 (m, 2H), 2.012.03 (m, 2H), 1.66-1.69 (m, 2H), 1.53-1.56 (m,
2H), 1.15-1.23 (m, 3H), and 1.11 (d, J = 6.5 Hz,
NMR (300 MHz, CDC13):
18 H)
13
C NMR (75 MHz, CDC13 ):
184.5, 125.9, 122.2, 31.6, 26.1, 23.4, 22.1, 21.2,
18.6, and 12.6
UV max (hexane):
231 nm (£ = 9200)
HRMS:
Calcd for C17 H3 0OSi:
Found:
186
278.2066
278.2067
I
C.
C0
L0
o
.-
iO
I-I
-c
V.
0.!
L
I
i
U
q&
L
(\J
0
-C
0
N
0
0a\
L
187
0
133
5 N2
N2
lb,
Sit-BuMe 2
171
133
1-Diazo-2-(1-cyclohexenyl)-1-(tert-butylsilyl)-2-ethenone
(171).
A 25-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 133 (0.139 g,
0.930 mmol) in 4 mL of a 50:50 solution of Et20-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (0.16 mL, 0.120 g, 0.930 mmol) was added dropwise over 1
min. Tert-butyldimethylsilyl trifluoromethanesulfonate (0.21 mL, 0.242 g, 0.914 mmol)
was added dropwise over 1 min and the resulting solution was stirred for 2 h while the ice-
water bath was allowed to warm to room temperature. The reaction mixture was filtered
through Celite with the aid of 5 mL of Et2O and the filtrate was concentrated to afford 0.25
g of an orange oil. Column chromatography on 10 g of silica gel (gradient elution with 510% EtOAc-hexane) provided 0.13 g (53%) of the silyldiazo ketone 171 as a yellow oil.
IR (thin film):
2935, 2867, 1704, 1675, 1578, 1550, 1249, 1221,
1119 and 1005 cm-l
1H
6.25-6.32 (m, 1H), 2.20-2.26 (m, 2H), 2.10-2.18
NMR (300 MHz, CDC13 ):
(m, 2H), 1.58-1.65 (m, 4H), 0.94 (s, 9H), and 0.22
(s, 6H)
188
K
to
D
m
I
.
o
1___7
'
i
O
0
0IN
cm
z
m
0
0
I-
cu
0
*(7
189
w
0
Si t-BuMe
2
N2
Sit-BuMe
183
2
171
C0
183
(l-Cyclohexenyl)-2-(tert-butyldimethylsilyl)-ketene
(181).
A 25-cm vycor tube (15 mm O.D., 13 mm I.D.) fitted with a rubber septum was
charged with a solution of silyldiazo ketone 171 (0.130 g, 0.490 mmol) in 5 mL of
benzene. A second rubber septum (inverted) was secured with wire to the tube to insure a
good seal, and the reaction mixture was degassed (three freeze-pump-thaw cycles at -196
°C, < 0.5 mm Hg) and then irradiated with 300 nm light for 4.5 h in a Rayonet reactor.
The resulting solution was concentrated to afford 0.11 g of an orange-brown oil. Column
chromatography on 10 g of silica gel (elution with hexane) provided 0.067 g (58%) of
ketene 183 as a viscous yellow oil.
IR (film):
2930, 2860, 2080, 1540, 1460, and 1250 cm- 1
1H
5.97-6.00 (m, 1H), 2.25-2.68 (m, 4H), 2.19-2.24
NMR (300 MHz, CDC13 ):
(m, 2H), 2.09-2.15 (m, 2H), 1.50 (s, 9H), and 0.75
(s, 6 H)
190
E
C.
.
0
.cu
I
m
--
o LO*
-
OD
o
o
*:4~
'r.G
r-
-a)
-C,
191
0
0
145
134
1-Diazo-3-methyl-3-penten-2-one
(134).
Reaction of methyl ketone 145 (3.04 g, 31.0 mmol) with LiHMDS (32.5 mmol)
and 2,2,2-trifluoroethyl trifluoroacetate (4.6 mL, 6.73 g, 34.4 mmol) in 60 ml of THF
according to the general procedure provided an orange oil which was then treated with H20
(0.56 mL, 0.56 g, 31.0 mmol), Et3 N (6.5 mL, 4.70 g, 46.5 mmol), and methanesulfonyl
azide (5.4 mL, 5.63 g, 46.5 mmol) in 35 mL of CH 3 CN at room temperature for 4 h to
yield 4.67 g of an orange oil. Column chromatography on 50 g of silica gel (elution with
20% EtOAc-hexane) provided 3.03 g (79%) of diazo ketone (134) as a yellow solid, mp
25.0-26.5 °C with spectral characteristics identical with those reported previously. 2
192
9
0
0
N2
N2
-
r
Si(iPr)3
134
161
l.-Diazo-3-methyl-1-(triisopropylsilyl)-3-penten-2-one
(161).
A 100-mL, three-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 134 (0.854 g,
6.87 mmol) in 40 mL of a 1:1 solution of Et 2 O-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (1.2 mL, 0.89 g, 6.87 mmol) was added dropwise over 1 min.
After 5 min, triisopropylsilyl trifluoromethanesulfonate (1.8 mL, 2.05 g, 6.70 mmol) was
added dropwise over 1 min and the resulting solution was stirred for 2.5 h while the icewater bath warmed to room temperature. The reaction mixture was filtered through Celite
with the aid of 10 mL of Et 2 O and the filtrate was concentrated to afford 2.09 g of an
orange oil. Column chromatography on 25 g of silica gel (elution with 5% EtOAc-hexane)
provided 1.71g (89%) of the silyldiazo ketone 161 as a yellow oil.
IR (CC14):
2940, 2860, 2065, 1610, 1460, 1383, 1340, 1279,
1209, and 1140 cm-
1H
6.13 (q, J = 6.6 Hz, 1H), 1.83 (s, 3H), 1.78 (d, J =
6.8 Hz, 3H), 1.37 (sept, J = 6.7 Hz, 3H), and 1.10
(d, J = 7.9 Hz, 18H)
NMR (300 MHz, CDC13):
1 3C
NMR (75 MHz, CDC13):
UV max (CH3CN):
HRMS:
196.9, 136.3, 129.9, 49.8, 18.4, 13.6, 13.1, and
11.5
292 ( = 3000), 238 (6200) and 221 (6300) nm
Calcd for C15H2 8 ON2Si:
Found:
193
280.1971
280.1963
o
CL
-u
0o
n
I
of
..
(01P
~~o
W
C.D
194
0
Si(i-Pr)
3
N2
Si(i.Pr) 3
161
175
3-Methyl-3-penten-2-triisopropylsilylketene
(175)
A solution of silyldiazo ketone 161 (0.76 g, 2.71 mmol) in 27 mL of benzene was
distributed evenly between two 25-cm vycor tubes (15 mm O.D., 13 mm I.D.) fitted with
rubber septa. A second rubber septum (inverted) was secured with wire to each tube to
insure a good seal, and the reaction mixtures were degassed (three freeze-pump-thaw
cycles at -196 °C, < 0.5 mm Hg) and then irradiated with 300 nm light for 4 h in a Rayonet
reactor. The resulting solutions were combined and concentrated to afford 0.71 g of a
yellow oil. Column chromatography on 30 g of silica gel (elution with hexane) provided
0.55 g (80%) of ketene 175 as a viscous yellow oil.
2941, 2881, 2081, 1461, 1381, 1291, 1181, 1071,
IR (film):
and 1021 cm- 1
1 H NMR
13C
(300 MHz, CDC13 ):
NMR (75 MHz, CDCl3):
5.18 (q, J = 6.7 Hz, 1H), 1.81 (s, 3H), 1.62 (d, J=
6.7 Hz, 3H), 1.15-1.27 (m, 3H), 1.10 (d, J = 6.4
Hz, 18H)
184.7, 123.9, 119.5, 22.4, 18.8, 18.6, 14.0, and
12.5
UV max (hexane):
229 nm ( = 9000)
HRMS
Calcd For C 1 5H 2 8 OSi:
Found:
195
252.1909
252.1907
I
i
I
-4
-0
-o
I
I
i
I
i
I
I
L
i
-u
-N
I/I
~~r
u,
In
i
1~~~~~~~~~~~~~~~~~~
--
tD
-cD
-0,
10IQ
rv
-o
0
0
4N 2
N2
16Et3
134
162
1-Diazo-3-methyl-1-triethylsilyl-3-penten-2-one
(162).
A 25-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 134 (0.201 g,
1.62 mmol) in 8 mL of a 50:50 solution of Et20-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (0.28 mL, 0.208 g, 1.61 mmol) was added dropwise over 1
min. After 5 min, triethylsilyl trifluoromethanesulfonate (0.36 mL, 0.43 g, 1.61 mmol)
was added slowly dropwise over 1 min and the resulting solution was stirred for 3 h while
the ice-water bath warmed to room temperature. The reaction mixture was filtered through
Celite with the aid of 5 mL of Et 2O and the filtrate was concentrated to afford an orange oil.
Column chromatography on 10 g of silica gel (gradient elution with 0-5% EtOAc-hexane)
provided 0.326 g (84%) of the silyldiazo ketone 162 as a yellow oil.
IR (CC):
2940, 2860, 2060, 1600, 1450, 1410, 1375, 1350,
1280, 1210, 1140, and 1000 cm-l
1H
6.13 (q, J = 6.8 Hz, 1H), 1.82 (s, 3H), 1.78 (d, J =
6.8 Hz, 3H), 0.98 (t, J = 7.8 Hz, 9H), and 0.77 (q,
J = 7.4 Hz, 6H)
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
196.4, 136.5, 130.3, 50.2, 13.6, 12.9, 7.1, and
2.9
UV max (CH3 CN):
219 ( = 6000), 240 (6100) and 292 (3100) nm
HRMS:
Calcd for C1 2 H2 2 N2 OSi (-N2):
Found:
197
210.1440
210.1433
-a
-6.
nr
v
-- l)~~
I
-N~
-Mp
t
c,'
z
v
1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ro
i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--ir
r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I
~~~~~~~~~~~~~I~~~~~~~~~~~~
i
I
I
e
c-m
198~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
r v~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SiEt
O
NN
2
C
o
176
162
3-Methyl-3-penten-2-triethylsilylketene
(176)
A solution of silyldiazo ketone 162 (0.419 g, 1.76 mmol) in 18 mL of benzene
was distributed evenly between two 25-cm vycor tubes (15 mm O.D., 13 mm I.D.) fitted
with rubber septa. A second rubber septum (inverted) was secured with wire to each tube
to insure a good seal, and the reaction mixtures were degassed (three freeze-pump-thaw
cycles at -196 C, < 0.5 mm Hg) and then irradiated with 300 nm light for 4 h in a Rayonet
reactor. The resulting solutions were combined and concentrated to afford a yellow-orange
oil. Column chromatography on 10 g of silica gel (elution with hexane) provided 0.27 g
(73%) of ketene 176 as a viscous yellow oil.
IR (film):
2940, 2860, 2065, 1445, 1380, 1300, 1235, 1280,
and 1010 cm- 1
1H
5.12 (q, J = 5.5 Hz, 1H), 1.80 (s, 3H), 1.62 (d, J=
5.9 Hz, 3H), 0.97 (t, J=7.9 Hz, 9H), and 0.70 (q, J
= 7.5 Hz, 6H)
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
184.0, 124.0, 117.7, 23.7, 18.6, 13.9, 7.2 and 3.9
UV max (CH3CN):
226 (e= 7500) and 352 (90) nm
HRMS:
Calcd for C12 H2 2 OSi:
Found:
199
210.1440
210.1437
I
a.
a
0
I;-------
---
__
n
I
.CU
-M
In_.V
O
o
I
I
I
____
*0
I
I
=Mlw
I
-D
-(D-
-C
mm
o
200
N2
135
1-Diazo-4-(2,6,6-trimethyl- 1-cyclohexen- 1-y)-3-buten-2-one
(135).
Reaction of 4-(2,6,6-trimethyl-l-cyclohexen-1-yl)-3-buten-2-one (2.1 mL, 2.00 g,
10.4 mmol) with LiHMDS (11.4 mmol) and 2,2,2-trifluoroethyl trifluoroacetate (1.7 mL,
2.45 g, 12.5 mmol) in 50 ml of THF according to the general procedure provided a yellow-
orange oil which was then treated with H20 (0.19 mL, 0.187 g, 10.4 mmol), Et 3N (2.2
mL, 1.58 g, 15.6 mmol), and methanesulfonyl
azide (1.8 mL, 1.89 g, 15.6 mmol) in 36
mi, of CH3 CN at room temperature for 3 h to yield 3.96 g of an orange oil. Column
chromatography on 50 g of silica gel (gradient elution with 0-10% EtOAc-hexane) provided
2.18 g (97%) of diazo ketone 135 as a yellow-orange oil.
IR (CCh):
3076, 2916, 2856, 2196, 2076, 1636, 1596, 1456,
1356 and 1296 cm- 1
1H
NMR (300 MHz, CDC13):
13C
NMR (75 MHz, CDC13):
7.32 (d, J = 15.8 Hz, 1H), 6.00 (d, J = 15.8 Hz,
1H), 5.36 (s, 1H), 2.06 (t, J = 6.0 Hz, 2H), 1.76
(s, 3H), 1.60-1.64 (m, 2H), 1.45-1.49 (m, 2H),
and 1.06 (s, 6H)
184.7, 140.1, 135.9, 135.8, 127.6, 55.3, 39.7,
34.0, 33.5, 28.7, 21.6, and 18.8
UV max (hexane):
300 nm ( = 18 000)
Elemental Analysis:
Calcd for C1 3 H18 0N2 :
Found:
201
C,
N,
C,
N,
71.52; H, 8.31;
12.84
71.81; H, 8.40;
12.69
pr
-.
'_o
0
0
o
Q
ao
0
-E
I
I$
`rc,~-1
'"-i~~~
0
Or
C
O
0
0
0In
L-1
0
CD
0
-1.o
0
0
b o
-o
-
-o
0
0
202
N2
:/-Pr)3
163
135
1-Diazo-1 -triisopropylsilyl-4-(2,6,6-trimethyl2-one
1-cyclohexen- 1-yl)-3-buten-
(163).
A 50-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic
stir bar and argon inlet adapter was charged with triisopropylsilyl
trifluoromethanesulfonate
(0.19 mL, 0.217 g, 0.707 mmol) in 17 mL of hexane and 10 mL
of Et2O and then cooled at 0 °C in an ice-water bath. A 10-mL, two-necked, roundbottomed flask equipped with a rubber septum and argon inlet adapter was charged with
diazo ketone 135 (0.150 g, 0.687 mmol) in 7 mL of Et 2 0 and cooled at 0 °C in an ice-
water bath while i-Pr2 EtN (0.12 mL, 0.088 g, 0.689 mmol) was added dropwise over 30
seconds. The diazo ketone solution was then transferred dropwise via cannula over 3 min
to the triflate solution.
The reaction mixture was stirred for 1.5 h allowing the bath to
warm to room temperature and then filtered through Celite with the aid of 10 mL of Et2O
and the filtrate was concentrated to afford 0.27 g of an orange oil.
Column
chromatography on 10 g of silica gel (gradient elution with 0-3% EtOAc-hexane) provided
0.23 g (87%) of the silyldiazo ketone 163 as an orange oil.
IR (film):
2942, 2967, 2057, 1632, 1601, 1461, and 1362
1H
7.35 (d, J = 15.5 Hz, 1H), 6.41 (d, J = 15.5 Hz,
1H) , 2.07 (t, J = 6.2 Hz, 2H), 1.79 (s, 3H), 1.601.64 (m, 2H), 1.46-1.50 (m, 2H), 1.37 (sept, J =
NMR (300 MHz, CDC13):
7.1 Hz, 3H), 1.13 (d, J = 7.3 Hz, 18H), and 1.08
(s, 6H)
203
13 C
NMR (75 MHz, CDC13 ):
188.5, 140.4, 136.3, 135.9, 124.6, 52.9, 39.8,
34.1, 33.7, 28.8, 21.8, 18.9, 18.4, and 11.7
UV max (hexane):
291 ( = 5200) and 272 (4500) nm
HRMS:
Calcd for C2 2 H 3 8 ON2Si:
Found:
204
374.2753
374.2755
-C-c.
I
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`i_
-
LI
z
X.
-I
oC'
I-
-r
-n
205
0
"'"'N
J
N2
Si(i-Pr) 3
163
178
4-(2,6,6-Trimethyl- 1-cyclohexen- 1-y)-3-buten-2-triisopropylsilylketene
(178).
A 20-cm vycor tube (8 mm O.D., 7 mm I.D.) fitted with a rubber septum was
charged with a solution of silyldiazo ketone 163 (0.23 g, 0.63 mmol) in 6 mL of benzene.
A second rubber septum (inverted) was secured with wire to insure a good seal, and the
reaction mixture was degassed (three freeze-pump-thaw cycles at -196 C, < 0.5 mm Hg)
and then irradiated with 300 nm light for 2 h in a Rayonet reactor. The resulting solution
was concentrated to afford 0.21 g of a yellow-brown oil. Column chromatography on 10 g
of silica gel (elution with hexane) provided 0.13 g (61%) of ketene 178 as a viscous
yellow oil.
IR (CC14):
2946, 2876, 2066, 1746, 1606, 1456 and 1346
cm-1
1H
NMR (300 MHz, CDC13):
5.90
1H),
1.61
3H),
(d, J = 15.8 Hz, 1H), 5.19 (d, J = 15.8 Hz,
1.97 (t, J = 6.1 Hz, 2H), 1.68 (s, 3H), 1.57(m, 2H), 1.42-1.46 (m, 2H), 1.16-1.23 (m,
1.11 (d, J = 6.2 Hz, 18H) and 0.98 (s, 6H)
13C NMR (75 MHz, CDC13):
185.3, 138.2, 128.3, 127.5, 120.4, 38.5, 34.3,
32.9, 28.8, 21.6, 19.4, 18.5, 15.6, and 12.0
UV max (hexane):
261 ( = 11 000), 206 (7100) nm
HRMS:
Calcd for C22H3 8 OSi:
Found:
206
346.2692
346.2695
I
a-
j
i
j
J
I--___
__4
I
-u
-I
l
IC
B
co
0
-in
l
-to
-T
--o
207
-o
136
(136).
I-Diazo-4-(2,6,6-trimethyl-2-cyclohexen-1-yl)-3-buten-2-one
Reaction of 4-(2,6,6-trimethyl- 1-cyclohexen-2-yl)-3-buten-2-one
(2.2 mL, 2.00 g,
10.4 mmol) with LiHMDS (11.4 mmol) and 2,2,2-trifluoroethyl trifluoroacetate (1.7 mL,
2.45 g, 12.5 mmol) in 30 ml of THF according to the general procedure provided a yellow-
orange oil which was then treated with H20 (0.19 mL, 0.19 g, 10.4 mmol), Et3N (2.2
mL, 1.57 g, 15.6 mmol), and methanesulfonyl
azide (1.8 mL, 1.89 g, 15.6 mmol) in 36
mL of CH3 CN at room temperature for 4.5 h to yield 3.32 g of a yellow-orange oil.
Column chromatography on 80 g of silica gel (gradient elution with 0-10% EtOAc-hexane)
provided 2.17 g (94%) of diazo ketone 136 as a yellow-orange oil.
IR (thin film):
3080, 2960, 2920, 2880, 2100, 1640, 1600, 1440
and 1360 cm-1
1H
NMR (300 MHz, CDC13 ):
13C
NMR (75 MHz, CDCl 3 ):
6.64 (dd, J = 15.1, 9.7 Hz, 1H), 5.96 (d, J =
15.5 Hz, 1H), 5.49 (br s, 1H), 5.33 (s, 1H), 2.25
(d, J = 9.4 Hz, 1H), 2.04 (s, 3H), 1.56 (s, 3H),
1.46 (dt, J = 13.4, 8.2 Hz, 1H), 1.21 (dt, J =
13.4, 4.8 Hz, 1H), 0.92 (s, 3H), and 0.85 (s, 3H)
184.5, 145.8, 131.9, 128.5, 122.6, 55.0, 54.1,
32.6, 31.2, 27.8, 26.9, 23.0, and 22.8
UV max (hexane):
293 (e = 8200) and 255 (15 800) nm
Elemental Analysis:
Calcd for C1 3 H18 0N2:
Found:
208
C,
N,
C,
N,
71.52; H, 8.31;
12.84
71.59; H, 8.40;
12.63
I
Z
'0.
0s
0to
0
To
-
.
0
-o
0
Lo
o
LO
A
0
-o
. --
S
_.0
0U
(D
S
-0
0n0
V
- 0D
O
O
209
PN2
J-Pr)3
136
164
l-Diazo-l-triisopropylsilyl-4-(2,6,6-trimethyl-2-cyclohexen2-one
1-yl)-3-buten-
(164).
A 10-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 136 (0.152 g,
0.690 mmol) in 3 mL of Et 2 0 and then cooled at 0 °C in an ice-water bath while i-Pr2EtN
(0.12 mL, 0.089 g, 0.69 mmol) was added dropwise over 1 min. Triisopropylsilyl
trifluoromethanesulfonate (0.19 mL, 0.217 g, 0.707 mmol) was added dropwise over 1
min and the resulting solution was stirred for 1 h while the ice-water bath warmed to room
temperature. The reaction mixture was filtered through Celite using 3 mL of Et2 0 and the
filtrate was concentrated to afford 0.28 g of an orange oil. Column chromatography on 10
g of silica gel (gradient elution with 0-2% EtOAc-hexane) provided 0.19 g (72%) of the
silyldiazo ketone 164 as a yellow oil.
IR (CC14):
2930, 2860, 2050, 1635, 1600, 1540, 1240, and
1200 cm-1
1H
NMR (300 MHz, CDC13 ):
13 C
NMR (75 MHz, CDCI 3 ):
6.74 (dd, J = 14.9, 9.8 Hz, 1H), 6.32 (d, J =
14.9 Hz, 1H), 5.48 (s, 1H), 2.27 (d, J = 9.7 Hz,
1H), 2.01-2.04 (m, 2H), 1.57 (s, 3H), 1.40-1.56
(m, 2H), 1.35 (sept, J = 5.7 Hz, 3H), 1.12 (d, J=
7.3 Hz, 18H), 0.93 (s, 3H) and 0.86 (s, 3H)
188.3, 146.1, 132.1, 125.4, 122.5, 54.1, 53.2,
32.6, 31.0, 27.9, 26.9, 23.1. 22.9, 18.5, and 11.7
UV max (hexane):
297 ( = 6100), 241 (10 000) and 225 (8800) nm
210
HRMS:
Calcd for C22H3 8 ON2 Si:
Found:
211
374.2753
374.2752
..
z
X.
I
S,
S
I
J-
IC
I
>
-e
Sl(Si(Pr)3
-A
~~d ~~~
164
"U"~
178
4- (2,6,6-Trimethylcyclo-2-hexen- 1-yl)-3-buten-2-triisopropylsilylketene
(178).
A 20-cm vycor tube (8 mm O.D., 7 mm I.D.) fitted with a rubber septum was
charged with a solution of silyldiazo ketone 164 (0.16 g, 0.43 mmol) in 4 mL benzene. A
second rubber septum (inverted) was secured with wire to each tube to insure a good seal,
and the reaction mixture was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5
mm Hg) and then irradiated with 300 nm light for 2.5 h in a Rayonet reactor. The resulting
solution was concentrated to afford 0.16 g of a red oil. Column chromatography on 10 g
of silica gel (elution with hexane) provided 0.12 g (81%) of ketene 178 as a viscous
yellow oil.
IR (CC14 ):
2916, 2856, 2066, 1536, and 1251 cm -1
1H
5.37 (br s, 1H), 5.19-5.32 (m, 2H), 2.10 (d, J =
7.6 Hz, 1H), 1.98 (br s, 2H), 1.57 (s, 3H), 1.39
NMR (300 MHz, CDC13 ):
(dt, J = 13.5, 7.8 Hz, 2H), 1.05-1.22 (m, 3H), 1.10
(d, J = 5.5 Hz, 18H), 0.87 (s, 3H), and 0.81 (s,
3H)
13C
NMR (75 MHz, CDC13 ):
185.3, 134.4, 131.4, 120.8, 117.9, 55.2, 32.5,
31.8, 27.6, 27.0, 23.1, 22.9, 18.4, 14.7, and 11.9
UV max (hexane):
243 nm (£ = 10 000)
HRMS:
Calcd for C22H3 8 OSi:
Found:
213
346.2692
346.2690
a
o
o
o
0
00
c
0
o
0
o
/P
ii~"*
V-
0C,
0
0
o
o
(0
214
0
0
Ph
h
-1
137
151
(137).
(E)-1-Diazo-6-phenyl-3-hexen-2-one
Reaction of methyl ketone 151 (0.687 g, 3.94 mmol) with LiHMDS (4.33 mmol)
and 2,2,2-trifluoroethyl trifluoroacetate (0.63 mL, 0.922 g, 4.70 mmol) in 28 ml of THF
according to the general procedure provided a yellow-orange oil which was then treated
with H20 (0.071 mL, 0.071 g, 3.94 mmol), Et 3 N (0.82 mL, 0.600 g, 5.91 mmol), and
methanesulfonyl azide (0.75 mL, 0.782 g, 6.45 mmol) in 14 mL of CH3 CN at room
temperature for 2.5 h to yield 1.43 g of an orange oil. Column chromatography on 50 g of
silica gel (gradient elution with 0-20% EtOAc-hexane) provided 0.68 g (86%) of diazo
ketone 137 as a yellow oil.
IR (CC14):
3024, 2924, 2099, 1655, 1614, and 1364 cm- 1
1H
7.26-7.32 (m, 2H), 7.16-7.23 (m, 3H), 6.84 (dt, J
= 15.4 Hz, 6.8 Hz, 1H), 5.99 (d, J = 15.5 Hz,
1H), 5.27 (s, 1H), 2.77 (t, J = 7.6 Hz, 2H), 2.52
NMR (300 MHz, CDC13):
(dt, J = 7.1 Hz, 7.3 Hz, 2H)
13C
NMR (75 MHz, CDC13):
184.6, 143.9, 140.8, 128.5, 128.4, 127.7, 126.2,
55.2, 34.5 and 34.0
UJVmax (hexane):
202 (e = 6400) and 247 (5700) nm
Elemental Analysis:
Calcd for C1 2 H12 0N2 :
Found:
215
C,
N,
C,
N,
71.97; H, 6.04;
13.99
71.72; H, 5.85;
13.85
-a
-n
oC
.c
LO
I)
VD
216
O
0
N2
Ph
PhN2
165
137
(E)-1-Diazo-6-phenyl-triisopropyl-3-hexen-2-one
Si(Pr)
3
(165).
A 250-mL, three-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, glass stopper, and argon inlet adapter was charged with triisopropyl
trifluoromethanesulfonate (0.67 mL, 0.760 g, 2.48 mmol) in 90 mL of a 50:40 solution of
hexane-Et20
then cooled at 0 °C in an ice-water bath. A 50-mL, two-necked, round-
bottomed flask equipped with a rubber septum, magnetic stir bar, and argon inlet adapter
was charged with diazo ketone 137 (0.496 g, 2.48 mmol) in 20 mL of ether and cooled at
0 °C in an ice-water bath while i-Pr2EtN (0.43 mL, 0.320 g, 2.48 mmol) was added
dropwise over 10 seconds. The diazo ketone solution was then transferred dropwise via
cannula over 5 min to the triflate solution. The reaction mixture was stirred for 1.5 h
allowing the ice-water bath to warm to room temperature and then filtered though Celite
using 5 mL of Et20 and the filtrate was concentrated to afford 2.80 g of an orange oil.
Column chromatography on silica gel (gradient elution with 0-50% benzene-hexane)
provided 0.34 g (38%) of the silyldiazo ketone 165 as an orange oil.
IR (thin film):
3015, 2935, 2855, 2040, 1650, 1605, 1460, and
1345 cm-
1H
NMR (300 MHz, CDCL3 ):
13C
NMR (75 MHz, CDC13):
1
7.17-7.32 (m, 5H), 6.91 (dt, J = 15.1, 7.5 Hz, 1H),
6.42 (d, J = 15.1 Hz, 1H), 2.79 (t, J = 7.7 Hz, 2H),
2.55 (dt, J = 7.0, 7.1 Hz, 2H), 1.35 (sept, J = 7.1
Hz, 3H) and 1.10 (d, J = 7.3 Hz, 18H)
188.2, 144.2, 140.9, 128.5, 128.3, 126.1, 124.4,
50.0, 34.5, 34.0, 18.4, and 11.6
217
UVNmax (hexane):
352 ( = 2500), 302 (6000) and 212 (16 000) nm
FIRMS:
Calcd for C2 1 H32 ON2 Si:
Found:
218
356.2284
356.2282
-x
-0a
-o
-cu
z
Z
D
-
a.
-m
m
I'
--
,"J
m
B
mq,,
1Y
0
Si(i-Pr 3 )
N2
Ph
165
165
Ph
Si(i-Pr 3 )
C
.0
179
(E)-6-Phenyl-3-hexen-2-triisopropylsilylketene
(179).
A solution of silyldiazo ketone 179 (0.29 g, 0.81 mmol) in 8 mL of benzene was
placed in a 20-cm vycor tube (8 mm O.D., 7 mm I.D.) fitted with a rubber septum. A
second rubber septum (inverted) was secured with wire to the tube to insure a good seal,
and the reaction mixture was degassed (three freeze-pump-thaw cycles at -196 C, < 0.5
mmHg) and then irradiated with 300 nm light for 2 h in a Rayonet reactor. The resulting
solution was concentrated to afford 0.30 g of a red oil. Column chromatography on 10 g
of silica gel (elution with hexane) provided 0.11 g (41%) of ketene 179 as a viscous
yellow oil.
IR (film):
3006, 2916, 2836, 2056, 1486, 1456 and 1386
cm-1
1H
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
7.15-7.29 (m, 5H), 5.49 (dt, J = 15.0, 6.9 Hz,
1H), 5.25 (d, J = 15.2 Hz, 1H), 2.67 (t, J = 7.6 Hz,
2H), 2.37 ( dt, J = 7.0,7.6 Hz, 2H), 1.06-1.18 (m,
3H) and 1.06 (d, J = 5.1 Hz, 18H)
184.9, 141.8, 129.7, 128.5, 128.3, 125.8, 117.5,
36.2, 35.0, 18.4, 14.6, and 11.9
UV max (hexane):
242 (£ = 8600) and 204 nm (16 000)
HRMS:
Calcd for C21 H32OSi:
Found:
220
328.2222
328.2221
I Do
C.
0.
M
-v
-i
, a-
0Q
/0
,,o
C)
U)
/
-W
-w
C-·-·L~~~~~~~~
''
0
O
N
Ph
2
Ph
N.
166
137
(E)- 1-Diazo-6- phenyl- 1-triethylsilyl-3-buten-2-one
N2
SiEt 3
(166).
A 25-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 137 (0.173 g,
0.860 mmol) in 6 mL of a 50:50 solution of Et20-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2 EtN (0.15 mL, 0.110 g, 0.860 mmol) was added dropwise over 1
min. Triethylsilyl trifluoromethanesulfonate (0.19 mL, 0.222 g, 0.840 mmol) was added
dropwise over 1 min, and the resulting solution was stirred for 2 h while the ice-water bath
warmed to room temperature. The reaction mixture was filtered through Celite using 5 mL
of Et2 O and the filtrate was concentrated to afford 0.29 g of an orange oil. Column
chromatography on 10 g of silica gel (gradient elution with 0-5% EtOAc-hexane) provided
0.14 g (52%) of the silyldiazo ketone 166 as a yellow oil.
IR (film):
2940, 2860, 2055, 1645, 1600, 1455, and 1380
cm-1
1H
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
7.22-7.30 (m, 2H), 7.17-7.20 (m, 3H), 6.82 (dt, J
= 6.9, 15.0 Hz, 1H), 6.33 (d, J = 15.0 Hz, 1H),
2.78 (t, J = 7.3 Hz, 2H), 2.53 (dt, J = 7.0, 6.9 Hz,
2H), 0.96 (t, J = 7.8 Hz, 9H), and 0.79 (q, J = 7.8
Hz, 6 H)
188.0, 144.1, 140.9, 128.4, 128.3, 126.1, 124.8,
53.5, 34.5, 33.9, 7.1, and 3.1
222
:g
-o
-to
z eVw.1
as
fl
V-
-U
-"'
223
O
Ph
SiEt
-JL
N2
166
Ph
SiEt 3
N
*0
180
(E)-6-Phenyl-3-buten-2-triethylsilylketene
(180).
A 20-cm vycor tube (8 mm O.D., 7 mm I.D.) fitted with a rubber septum was
charged with a solution of silyldiazo ketone 166 (0.140 g, 0.450 mmol) in 4 mL of
benzene. A second rubber septum (inverted) was secured with wire to the tube to insure a
good seal, and the reaction mixture was degassed (three freeze-pump-thaw cycles at -196
°C, < 0.5 mm Hg) and then irradiated with 300 nm light for 2.5 h in a Rayonet reactor.
The resulting solution was concentrated to afford 0.15 g of a dark red oil. Column
chromatography on 10 g of silica gel (elution with hexane) provided 0.028 g (20%) of
ketene 180 as a viscous yellow oil.
IR (thin film):
3023, 2995, 2867, 2083, 1730, 1633, and 1450
cm-
1H
7.24-7.36 (m, 2H), 7.15-7.17 (m, 3H), 5.32-5.50
(m, 2H), 2.65 (t, J = 7.6 Hz, 2H), 2.37 (dt, J =
7.3, 7.6 Hz, 2H), 0.95 (t, J = 7.5 Hz, 9 H), and
0.65 (q, J = 7.5 Hz, 6H)
NMR (300 MHz, CDC13):
224
i
c'
U
U/o
0
-
-
-
0
0
N~~phA~2
,N2
170
141
(E)-l-Diazo-4-phenyl-l-triisopropylsilyl-3-buten-2-one
Si(-Pr) 3
(170).
A 250-mL, three-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, glass stopper, and argon inlet adapter was charged with a solution of
tri-isopropyl trifluoromethanesulfonate
(0.78 mL, 0.889 g, 2.90 mmol) in 115 mL of Et2O
then cooled at 0 C in an ice-water bath. A 50-mL, two-necked, round-bottomed
flask
equipped with a rubber septum, magnetic stir bar, and argon inlet adapter was charged with
diazo ketone 141 (0.509 g, 2.90 mmol) in 23 mL of Et 2 O and cooled at 0 C in an ice-
water bath while i-Pr2 EtN (0.51 mL, 0.380 g, 2.90 mmol) was added dropwise over 10
seconds. The solution of diazo ketone was then transferred dropwise via cannula over 4
min to the solution of triflate. The reaction mixture was stirred for 15 min while the ice-
water bath warmed to room temperature and then filtered through Celite using 10 mL of
Et20 and the filtrate was concentrated to afford 1.08 g of a red-orange oil. Column
chromatography on 40 g silica gel (gradient elution with 0-5 % EtOAc-hexane) provided
0.37 g (39%) of the silyldiazo ketone 170 as an orange oil.
IR (CC14):
2940, 2860, 2058, 1640, 1600, 1460, 1330, 1280,
1210, and 1180 cm-1
1H
7.64 (d, J = 15.3 Hz, 1H), 7.54-7.57 (m, 2H),
7.37-7.40 (m, 3H), 7.04 (d, J = 15.4 Hz, 1H),
1.40 (sept, J = 7.2 Hz, 3H), and 1.14 (d, J = 7.4
Hz, 18H)
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
UV max (CH 3CN):
187.9, 141.1, 134.7, 130.2, 128.9, 128.2, 120.4,
54.0, 18.4, and 11.6
351 ( = 3100) and 223 (6900) nm
226
HRMS:
Calcd For C19H28 ON2 Si (-N2):
Found:
227
300.1909
300.1901
-0a
s
~~~~
~~~~~~~~~~~~~j
-c
z
J
.
-
-(D
-
-rn
-a'
228
0
Ph~,N
Si(i-Prh
2
Ph--~C
Si(i-P)r3
170
182
(E)-4-Phenyl-3-buten-2-triisopropylsilylketene
(182).
A solution of silyldiazo ketone 170 (0.37 g, 1.13 mmol) in 12 mL of benzene was
placed in a 25-cm vycor tube (15 mm O.D., 13 mm I.D.) fitted with a rubber septum. A
second rubber septum (inverted) was secured with wire to the tube to insure a good seal,
and the reaction mixture was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5
mm Hg) and then irradiated with 300 nm light for 3 h in a Rayonet reactor. The resulting
solution was concentrated to afford 0.25 g of a red oil. Column chromatography on 10 g
of silica gel (elution with hexane) provided 0.12 g (35%) of ketene 182 as a viscous
yellow oil.
IR (thin film):
2934, 2869, 2083, 1604, 1460, 1375, 1320, and
1010 cm- 1
1H
NMR (300 MHz, CDC13):
7.15-7.27 (m, 5H), 6.40 (d, J = 15.8 Hz, 1H), 6.01
(d, J = 15.8 Hz, 1H), 1.24 (sept, J = 6.3 Hz, 3H),
and 1.13 (d, J = 6.6 Hz, 18H)
l3C NMR (75 MHz, (CD3 )2 SO):
HRMS:
184.5, 137.4, 128.6, 128.4, 126.6, 125.2, 117.4,
18.2, 16.5, and 11.1
Calcd For C 19 H2 8OSi:
Found:
229
300.1909
300.1904
E
I
0
-1
.4
-
-
-I
N~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~N
a..
o/
.4.
..
Ix V-
~ 0 ~~~~~~~~~~~~~~~~~~~~~~~.
~~~
(O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
F
I
L- I-
I
--
I~~~r
-
N
,
_ C
230
O
,N2
153
138
1-Diazo-3-cyclohexyl-3-buten-2-one
(138).
Reaction of methyl ketone 153 (0.216 g, 1.56 mmol) with LiHMDS (1.75 mmol)
and 2,2,2-trifluoroethyl trifluoroacetate (0.26 mL, 0.381 g, 1.91 mmol) in 5 ml of THF
according to the general procedure provided a yellow oil which was then treated with H 2 0
(0.029 mL, 0.029 g, 1.59 mmol), Et 3 N (0.33 mL, 0.240 g, 2.39 mmol), and
methanesulfonyl azide (0.28 mL, 0.292 g, 2.39 mmol) in 6.5 mL of CH3 CN at room
temperature for 2 h to yield 0.28 g of a yellow oil. Column chromatography on 10 g of
silica gel (gradient elution with 0-10 % EtOAc-hexane) provided 0.170 g (66%) of diazo
ketone (138) as a yellow oil.
IR (CC14):
2926, 2831, 2096, 1646, 1619, 1601, 1551, 1492,
and 1446 cm-1
1H
5.68 (s, 1H), 5.19 (s, 1H), 2.87 (br s, 2H), 2.16 (t,
J = 5.9 Hz, 2H), and 1.61-1.68 (m, 6H)
NMR (300 MHz, CDC13):
13 C
NMR (75 MHz, CDC13):
186.0, 161.0, 118.7, 56.4, 37.9, 30.1, 28.7, 27.8,
and 26.2
231
"x
-a
-o
-U
-
-
iI
-mu
-3~~~~~~~
"N
I
a
£
-L0
j
z\
IS
A
m
M,
232
N2
)3
-
167
138
1-Diazo-4-cyclohexyl-
1-triisopropylsilyl-3-buten-2-one
(167).
A 25-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 138 (0.110 g,
0.670 mmol) in 6 mL of a 50:50 solution of Et2O-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (0.11 mL, 0.0816 g, 0.632 mmol) was added dropwise over 1
min. After 5 min, triisopropylsilyl trifluoromethanesulfonate (0.18 mL, 0.200 g, 0.660
mmol) was added dropwise over 1 min and the resulting solution was stirred for 2 h while
the ice-water bath warmed to room temperature. The reaction mixture was filtered through
Celite with the aid of 5 mL of Et2 O and the filtrated was concentrated to afford 0.33 g of a
yellow oil. Column chromatography on 10 g of silica gel (gradient elution with 0-5%
EtOAc-hexane) provided 0.200 g (93%) of the silyldiazo ketone 167 as a yellow oil.
IR (thin film):
2954, 2884, 2074, 1654, 1624, 1474, 1404, 1284,
1254, and 1184 cm-1
1H
6.06 (s, 1H), 2.68 (t, J = 5.9 Hz, 2H), 2.18 (t, J =
5.8 Hz, 2H), 1.59-1.67 (m, 6H), 1.36 (sept, J = 7.1
Hz, 3H), and 1.11 (d, J = 7.3 Hz, 18H)
NMR (300 MHz, CDC13):
13C
NMR (75 MHz, CDC13):
190.9, 158.0, 117.9, 53.1, 37.6, 30.1, 28.6, 27.8,
26.3, 18.4, and 11.7
233
.0,.
- o
-,w.
-------·
--
I
i-
I
-- - _
-cu
Ii
-m
z
.
I
.4i
u,~
-U
234
-o
140
(E)- 1-Diazo-3-penten-2-one
(140).
Reaction of trans-3-penten-2-one ( 0.58 mL, 0.500 g, 5.94 mmol) with LiHMDS
(6.54 mmol) and 2,2,2-trifluoroethyl
trifluoroacetate (0.95 mL, 1.39 g, 7.09 mmol) in 15
ml of THF according to the general procedure provided an orange-brown oil which was
then treated with H20 (0.11 mL, 0.110 g, 6.10 mmol), Et3N (1.2 mL, 0.871 g, 8.61
rnmol), and methanesulfonyl
azide (1.0 mL, 1.04 g, 8.60 mmol) in 20 mL of CH 3 CN at
room temperature for 4 h to yield 0.47 g of an orange oil. Column chromatography on 10
g of silica gel (gradient elution with 0-20% EtOAc-hexane) provided 0.27 g (41%) of diazo
ketone (140) as a yellow oil.
IR (CC14):
3120, 2950, 2090, 1650, 1615, 1540, 1440, 1360,
1300, and 1150 cm-l
1H
6.83 (dq, J = 15.0, 7.0 Hz, 1H), 6.00 (d, J = 15.2
Hz, 1H), 5.29 (s, 1H), and 1.89 (d, J = 7.0 Hz,
3H)
NMR (300 MHz, CDC13 ):
235
-
I
xI
c,-
I
i
BC-C-'
I-e
z
-r-
-C,
236
-n---
o
o
N2
N2
'N
140
169
(E)-l-Diazo-l-triisopropylsilyl-3-penten-2-one
Si(iPr)
3
(169).
A 25-mL, two-necked, round-bottomed flask equipped with a rubber septum,
magnetic stir bar, and argon inlet adapter was charged with diazo ketone 140 (0.159 g,
1.44 mmol) in 14 mL of a 50:50 solution of Et20-hexane and then cooled at 0 °C in an ice-
water bath while i-Pr2EtN (0.26 mL, 0.193 g, 1.49 mmol) was added dropwise over 1
min. After 5 min, triisopropylsilyl trifluoromethanesulfonate (0.39 mL, 0.445 g, 1.45
mmol) was added dropwise over 1 min and the resulting solution was stirred for 3 h while
the ice-water bath warmed to room temperature. The reaction mixture was filtered through
Celite with the aid of 5 mL of Et20 and the filtrate was concentrated to afford 0.42 g of an
orange oil. Column chromatography on 10 g of silica gel (gradient elution with 0-5%
EtOAc-hexane) provided 0.060 g (16%) of the silyldiazo ketone 169 as a yellow oil:
IR (CC14):
2950, 2880, 2080, 1660, 1610, 1550, 1470, 1389,
1300, 1220, and 1125 cm- 1
1H
6.89 (dq, J = 15.3, 7.2 Hz, 1H), 6.46 (d, J = 15.0
Hz, 1H), 1.90 (d, J = 7.0 Hz, 3H), 1.37 (sept, J=
NMR (300 MHz, CDC13):
7.6 Hz, 3H), and 1.11 (d, J = 7.3, 18 H)
13C NMR (75 MHz, CDC13):
188.3, 140.6, 125.5, 50.0, 18.4, 17.9, and 11.6
237
x
I
a-
i
i
I
l
-
i
i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
F~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
eb
e8
z2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
l.~~~~~~~~~~~~~~~~~~~~~~~~~~
Al~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Ln
-r
1
I
--
0
-1-1
I
-M
I
I
I
_cn
o
238
e-
O
Ph
O
PhIO
N
2
139
158
(E)-l-Diazo-5-benzyloxy-3-penten-2-one
(139).
Reaction of methyl ketone 158 (0.274 g, 1.44 mmol) with LiHMDS (1.58 mmol)
and 2,2,2-trifluoroethyl trifluoroacetate (0.23 mL, 0.339 g, 1.73 mmol) in 6 ml of THF
according to the general procedure provided a yellow-orange oil which was then treated
with H2 0 (0.025 mL, 0.025 g, 1.39 mmol), Et3 N (0.30 mL, 0.219 g, 2.16 mmol), and
methanesulfonyl azide (0.25 mL, 0.262 g, 2.16 mmol) in 10 mL of CH3 CN at room
temperature for 3 h to yield 0.30 g of an orange oil. Column chromatography on 30 g of
silica gel (gradient elution with 0-20% EtOAc-hexane) provided 0.23 g (74%) of diazo
ketone 139 as a yellow oil.
IR (thin film):
3090, 2860, 2110, 1660, 1600, 1590, 1455, and
1360 cm - 1
1H
7.22-7.41 (m, 5H), 6.83 (dt, J = 15.5, 4.1 Hz,
1H), 6.24 (d, J = 16.1 Hz, 1H), 5.32 (s, 1H), 4.58
NMR (300 MHz, CDC13):
(s, 2H), and 4.10 (dd, J = 2.0, 4.0 Hz, 2H)
239
E
-a
I0cu
OIT
k
_
cm~~~~~~~i
-i\)~~~~~~~~~~~~~~~~~~~~~~
0"J
1o,~~~~~~~~~~~~~~~~~~~~
_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
A&
X
_ L
_V
2
1
Q-
I
r
-(D
I
-rI'--
1
C)
I0
--
240
C
(i-Pr) 3Si
(iPr)
3S
C02e
175
227
236
Dimethyl-5,6-dimethyl-3-hydroxy-4-triisopropylsilyl
phthalate (236).
A flame-dried, threaded Pyrex tube (20 mm O.D., 16 mm I.D.) was charged with a
solution of silylketene 175 (0.092 g, 0.370 mmol) and dimethyl acetylenedicarboxylate
(0.045 mL, 0.052 g, 0.370 mmol) in 0.4 mL of toluene and sealed with a rubber septum.
The reaction mixture was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5 mm
Hg) and then tightly sealed with a teflon cap. The reaction mixture was heated at 150 °C
for 24 h and then transferred to a round-bottomed flask using 1 mL of benzene and
concentrated to afford 0.20 g of an orange oil. Column chromatography on 10 g of silica
gel (gradient elution with 0-70% benzene-hexane) provided 0.14 g (95%) of 236 as white
crystals: mp 73-75 C.
IR (CC14):
2950, 2870, 1740, 1670, 1445, 1350, 1330, 1225,
and 1050 cm-1
1H
11.57 (s, 1H), 3.90 (s, 3H), 3.88 (s, 3H), 2.37 (s,
3H), 2.10 (s, 3H), 1.58 (sept, J = 7.5 Hz, 3H), and
NMR (300 MHz, CDC13):
1.08 (d, J = 7.2 Hz, 18H)
13C NMR (75 MHz, CDCl 3):
170.4, 170.2, 165.7, 135.2, 128.3, 126.5, 124.5,
104.9, 52.6, 52.2, 22.1, 19.3, 16.9, and 13.6
UV max (CH 3 CN):
322 (E = 9700), 257 (9900), and 220 (32 000) nm
Elemental Analysis:
Calcd for C21 H3 4 05Si:
Found:
241
C, 63.92;
H, 8.69
C, 63.88; H, 8.67
0
0
S
O
O
aO
a.
0
0
C
L
0
-o
te
Cl)
-o
-01
-o
0
K.
tD
O
K
L/+tL
OAc
(i-Pr) 3 S
:02Me
Me
CO2Me
2 Me
236
237
Dimethyl 3-acetoxy-5,6-dimethyl-4-triisopropylsilyl phthalate (237).
A 5-mL, two-necked, round-bottomed flask equipped with a rubber septum and
argon inlet adapter was charged with a solution of phenol 236 (0.0629 g, 0.160 mmol).
DMAP (0.023 g, 0.19 mmol) and acetic anhydride (0.021 mL, 0.023 g, 0.21 mmol) in 2
mL of dichloromethane were then added at room temperature. The reaction mixture was
stirred for 18 h at room temperature and 1 mL of methanol was then added. The reaction
mixture was then transferred to a round-bottomed flask using
mL of dichloromethane and
concentrated to afford a brown-green oil. The oil was then dissolved in 2 mL of Et20 and
washed with 5 mL of 5% HCI, 5 mL of NaHCO3, dried over MgSO4, filtered, and
concentrated to afford 0.056 g of a brown oil. Column chromatography on 5 g of silica gel
(gradient elution with 0-20% EtOAc-hexane) provided 0.047 g (67%) of 237 as a yellow
oil.
IR (CC14):
2950, 2880, 1770, 1735, 1550, 1435, 1390, 1365,
1290, 1270, 1210, 1190, 1050, and 1010 cm - 1
1H
3.89 (s, 3H), 3.78 (s, 3H), 2.40 (s, 3H), 2.25
(s,3H), 2.22 (s, 3H), 1.55 (sept, J = 7.2 Hz, 3H),
NMR (300 MHz, CDC13):
and 1.10 (d, J = 7.5 Hz, 18H)
13 C
NMR (75 MHz, CDC13):
170.0, 169.3, 169.2, 166.0, 132.3, 127.5, 125.1,
123.8, 107.6, 52.5, 52.4, 23.0, 21.6, 19.4, 17.3,
and 14.0
243
-I
-o
-Cu
a
cm
244
OH
(i-Pr)3 Si
OH
2Me
2Me
02Me
) 2 Me
238
236
Dimethyl 5,6-dimethyl-3-hydroxy
phthalate (238).
A 10-mL, two-necked, round-bottomed flask equipped with a rubber septum and
argon inlet adapter was charged with a solution of phthalate 236 (0.200 g, 0.507 mmol)
and 0.6 mL of dichloromethane.
Trifluoroacetic acid (0.78 mL, 1.16 g, 10.1 mmol) was
added dropwise to the reaction mixture over 1 min and the resulting solution was stirred at
room temperature for 2 h. The reaction mixture was then transferred using 0.5 mL of
dichloromethane to a round-bottomed flask and concentrated to afford 0.29 g of a brown
oil. Column chromatography on 15 g of silica gel (gradient elution with 0-10% EtOAchexane) provided 0.092 g (76%) of phthalate 238 as a colorless oil.
IR (CHC13):
3700, 3640, 3040, 2990, 1730, 1660, 1525, 1480,
1450, 1430, and 1340 cm- 1
1H
10.76 (s, 1H), 6.84 (s, 1H), 3.88 (s, 6H), 2.25 (s,
3H), and 2.07 (s, 3H)
NMR (300 MHz, CDC13 ):
245
ICL
ci
I
-ir
C)
0
0
o
8
o
0
-
0
.4
Cu
cli
.4
246
(i-Pr)Si
AAA
C
+
CH2CH2Ph
179
·· _
( Pr)
''U
Vrl~~~~~~~~-
II
CO2Me
242
227
Dimethyl 6-(2-phenylethyl)-3-hydroxy-4-triisopropylsilyl phthalate (242).
A flame-dried, threaded Pyrex tube (20 mm O.D., 16 mm I.D.) was charged with a
solution of silylketene 179 (0.090 g, 0.27 mmol) and dimethyl acetylenedicarboxylate
(0.050 mL, 0.058 g, 0.41 mmol) in 0.3 mL of toluene and sealed with a rubber septum.
The reaction mixure was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5 mm
Hg) and then tightly sealed with a teflon cap. The reaction mixture was heated at 150 °C
for 3 h and then transferred to a round-bottomed flask using 1 mL of benzene and
concentrated to afford 0.12 g of a brown oil. Column chromatography on 10 g of silica gel
(gradient elution with 0-50% CH2C12-petroleumether) provided 0.079 g (61%) of 242 as
a white solid: mp 77.5-79.0 °C.
IR (CC14):
2943, 2860, 1740, 1670, 1585, 1440, 1400, 1340,
1210, 1135, and 1030 cm -1
1H
11.19 (s, 1H), 7.12-7.29 (m, 6H), 3.91 (s, 3H),
3.90 (s, 3H), 2.79-2.85 (m, 4H), 1.43 (sept, J=
NMR (300 MHz, CDC13):
7.6 Hz, 3H), and 1.04 (d, J = 7.6 Hz, 18 H)
13C
NMR (75 MHz, CDC13):
170.0, 169.6, 165.0, 144.6, 141.1, 134.9, 128.5,
128.5, 128.4, 126.3, 126.0, 107.7, 52.8, 52.3,
37.7, 35.0, 18.8, and 11.5
IUVmax (CH 3 CN):
310 ( = 5900), 317 (7500), and 219 (32 000) nm
Elemental Analysis:
Calcd for C27H3 805Si:
Found:
247
C, 68.90; H, 8.14
C, 68.80; H, 8.17
-
-cu
n
c
C
-C
-L
L
-o
. C
- I
LI
248
IL
CO2 Me
(i-Pr)3
2 Me
+
II
02 Me
C0 2Me
181
227
243
1,2-Dimethyl 5,6,7,8-Tetrahydro-4-triisopropylsilyl-3-napthol
dicarboxylate (243).
A flame-dried, threaded Pyrex tube (20 mm O.D., 16 mm I.D.) was charged with a
solution of silylketene 181 (0.154 g, 0.55 mmol) and dimethyl acetylenedicarboxylate
(0.066 mL, 0.076 g, 0.54 mmol) in 0.4 mL of toluene and sealed with a rubber septum.
The reaction mixure was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5 mm
Hg) and then tightly sealed with a teflon cap. The reaction mixture was heated at 150 °C
for 22 h and then transferred to a round-bottomed flask using 1 mL of benzene and
concentrated to afford 0.29 g of a brown oil. Column chromatography on 10 g of silica gel
(gradient elution with 0-80% benzene-hexane) provided 0.17 g (73%) of 243 as a light
yellow oil.
IR (CC):
2950, 2870, 1740, 1670, 1575, 1440, 1380, 1325,
and 1212 cm- 1
1H
11.45 (s, 1H), 3.90 (s, 3H), 3.89 (s, 3H), 2.86 (t, J
= 5.9 Hz, 3H), 2.59 (t, J = 6.4 Hz, 3H), 1.60-1.71
(m, 4H), 1.54 (sept, J = 7.3 Hz, 3H), and 1.06 (d, J
= 7.3 Hz, 18H)
NMR (300 MHz, CDC13 ):
13C
NMR (75 MHz, CDC13 ):
170.0, 169.9, 165.0, 155.1, 135.7, 125.6, 125.3,
105.1, 52.8, 52.3, 31.8, 26.3, 22.1, 21.7, 19.5,
and 13.8
249
I
o
-C
-0
qNu
-ID~1
LJU
A
_
('Pr)
3
C
Si
U
NC
CO2 Et
CH3
(Pr) S
3
4.
ri
175
244
245
6-Cyano-6-ethylcarboxylate-3,4-dimeth yl-2-triisopropylsilyl-2-cyclohexen1-one
(244,245).
A 10-mL, two-necked round-bottomed flask equipped with a rubber septum and
argon inlet adapter was charged with a solution of silylketene 175 (0.201 g, 0.800 mmol)
in toluene while ethyl cyanoacrylate (0.11 mL, 0.12 g, 0.96 mmol) was added dropwise at
room temperature over 0.5 min. The reaction mixture was stirred at room temperature for
24 h and then transferred to a round-bottomed flask using 1 mL of benzene and
concentrated to afford 0.39 g of a light yellow oil. Column chromatography on 10 g of
silica gel (gradient elution with 0-20% EtOAc-hexane) provided 0.113 g (38%) of 244, mp
81..5-82 °C and 0.625 g (20%) of 245, mp 65-66.5 °C, and 0.120 g (40%) of a mixture of
244 and 245 as white solids (total yield, 98%).
IR (CC14):
For the major diastereomer 244:
2940, 2870, 2240, 1750, 1690, 1570, 1450, 1370,
1240, 1215, 1160, 1110, 1070, and 1020 cm-l
For the minor diastereomer 245:
2950, 2870, 1755, 1695, 1550, 1460, 1370, 1250,
1160, and 1075 cm-1
1HiNMR (300 MHz, CDC13):
For the major diastereomer 244:
4.29 (dq, J = 10.8, 7.1 Hz, 1H), 4.18 (dq, J =
10.7, 7.1 Hz, 1H), 2.89 (dd, J = 13.9, 6.0 Hz,
1H), 2.72 (app q, J = 6.7 Hz, 1H), 2.15 (dd, J=
13.9, 6.4 Hz, 1H), 2.08 (s, 3H), 1.44 (sept, J = 7.5
Hz, 3H), 1.35 (d, J = 7.1 Hz, 3H), 1.31 (t, J = 7.1
Hz, 3H), 1.06 (d, J = 7.5 Hz, 12H), and 1.04 (d, J
= 7.4 Hz, 6H)
251
For the minor diastereomer 245:
4.39 (dq, J = 10.8, 7.2 Hz, 1H), 4.30 (dq, J =
12.0, 7.2 Hz, 1H), 2.81 (ddq, J = 5.4, 10.7, 7.0
Hz, 1H), 2.52 (dd, J = 14.2, 5.4 Hz, 1H), 2.25
(dd, J = 14.2, 10.7 Hz, 1H), 2.10 (s, 3H), 1.43
(sept, J = 7.5 Hz, 3H), 1.34 (t, J = 7.1 Hz, 3H),
1.27 (d, J = 7.0 Hz, 3H), and 1.08 (app t, J = 7.5
Hz, 18H)
13C
NMR (75 MHz, CDC13):
For the major diastereomer 244:
189.7, 174.6, 165.0, 132.1, 117.4, 63.9, 55.1,
37.3, 35.9, 24.6, 21.0, 19.4, 19.3, 14.2, 13.1, and
13.0
For the minor diastereomer 245:
191.0, 172.5, 164.9, 132.3, 115.9, 63.1, 55.0,
37.1, 36.0, 23.4, 20.2, 19.1, 18.9, 14.0, and 12.7
Elemental Analysis:
Calcd for C2 1 H3 503SiN:
C, 66.79; H, 9.34;
N, 3.71
Found for the major diastereomer 244:
C, 66.57; H, 9.42;
N, 3.62
Found for the minor diastereomer 245:
C, 66.86; H, 9.41;
N, 3.62
252
0
(iPr) 3Si.
'CN
244
,to........ 6.......
.... j .... I.... 41 - 11-111 11.... , .. I .... I ... ,I .... I .... I.... 4b; .
(i-Pr)3Si.
245
I-r
1,911
to
I1
9
11
. . .
a
.
. 1.1|11
..
I
d
.
.
. I.
i
.
. . .
. . .
253
. . .
-
. . .
. .
f
.
.
.
.
.
.
I
..
.lil
OPP
i
I I
..
O
O
(/Pr) 3Si.
NC
CO2 Et
2CH2CH 3
(i-Pr) 3 Si,
III
0,
246a,b
177
6-Cyano-6-ethylcarboxylate-2-triisopropylsilyl-4-(2,6,6-trimethyl-
cyclohexen-1-yl)-2-cyclohexen-l-one
1-
(246a,b).
A 10-mL, two-necked round-bottomed flask equipped with a rubber septum and
argon inlet adapter was charged with a solution of silylketene 177 (0.209 g, 0.600 mmol)
in toluene while ethyl cyanoacrylate (0.11 mL, 0.118 g, 0.940 mmol) was added dropwise
at room temperature over 0.25 min. Five additional (0.05 mL, 0.054 g, 0.43 mmol)
portions of ethyl cyanoacrylate were added at intervals of -15 h. After 93 h, the reaction
mixture was transferred to a round-bottomed flask using 2 mL of benzene-acetone and
concentrated to afford 0.75 g of a light yellow oil. Column chromatography on 10 g of
silica gel (gradient elution with 0-3% EtOAc-hexane) provided 0.12 g (43%; 86% yield
based on recovered starting material) of a 4:1 mixture of diastereomers 246a and 246b as
a yellow oil.
IR (CC14):
For the mixture of diastereomers:
2940, 2860, 1740, 1680, 1540, 1460, 1430, 1360,
1230, 1100, 1060, and 1001
1H
For the major diastereomer 246a:
7.07 (s, 1H), 4.29 (dq, J = 9.0, 7.1 Hz, 1H), 4.26
(dq, J = 10.8, 7.2 Hz, 1H), 3.53 (dd, J = 11.2, 3.1
Hz, 1H), 2.84 (dd, J = 14.0, 5.0 Hz, 1H), 2.66
(dd, J = 14.0, 11.7 Hz, 1H), 1.89-1.96 (m, 2H),
1.57 (s, 3H), 1.43-1.65 (m, 4H), 1.26-1.41 (m,
3H), 1.33 (t, J = 7.1 Hz, 3H), 1.05 (s, 6H), and
1.04 (d, J = 7.5 Hz, 18 H)
NMR (300 MHz, CDC13):
254
HRMS:
Calcd for C28H4 5 0 3NSi:
Found:
255
471.3169
471.3164
I0.
a
--0
-.
.D
ci
N
a0.
.4.
S,
-
,.o
D-
-fl
256
(iPr) 3 S
C
175
NO2
249a,b
248
cis-6-Nitro-3,4-trimethyl-2-triisopropylsilyl-2-cyclohexen-1-one
and
trans-6-Nitro-3,4-trimethyl-2-triisopropylsilyl-2-cyclohexen- l-one
(249a,b).
A 10-mL, two-necked, round-bottomed flask equipped with a rubber septum and
argon inlet adapter was charged with a solution of silylketene 175 (0.208 g, 0.822 mmol),
two crystals of BHT, and 0.8 mL of benzene. A solution of nitroethylenel3 3 (0.99 M in
benzene, 0.83 mL, 0.82 mmol) was added, and the resulting mixture was stirred at room
temperature. Two additional portions of nitroethylene solution were added (0.99 M in
benzene, 0.83 mL, 0.82 mmol) after 12 h and 20 h. The reaction mixture was stirred for a
total of 39 h and then transferred to a round-bottomed flask using 1 mL of benzene and
concentrated to afford 0.29 g of a light yellow oil. Column chromatography on 30 g of
silica gel (gradient elution with 0-10% EtOAc-hexane) provided 0.23 g (85%) of a 2:1
mixture of diastereomers 249a and 249b as a yellow oil.
IR (CC14):
For the mixture of diastereomers 249a,b:
2930, 2850, 1735, 1670, 1550, 1450, 1360, 1280,
1240, and 1200 cm-l1
1H
NMR (300 MHz, CDC13):
For the major diastereomer 249a:
5.42 (dd, J =4.2, 12.5 Hz, 1H), 2.88 (dd, J = 5.7,
13.1 Hz, 1H), 2.53-2.75 (m, 1H), 2.27-2.41 (m,
1H), 2.13 (s, 3H), 1.43 (sept, J = 7.6 Hz, 3H),
1.34 (d, J = 7.2 Hz, 3H), and 1.05 (d, J = 7.4 Hz,
18 H)
For the minor diastereomer 249b:
5.37 (dd, J = 4.7, 12.4 Hz, 1H), 2.80 (dd, J = 6.0,
13.7 Hz, 1H), 2.53-2.73 (m, 1H), 2.27-2.41 (m,
1H), 2.08 (s, 3H), 1.54 (sept, J = 7.5 Hz, 3H),
1.29 (d, J = 6.9 Hz, 3H), and 1.08 (d, J = 7.4 Hz,
18H)
257
13 C
NMR (75 MHz, CDC13 ):
For the mixture of diastereomers 249a and 249b:
191.9, 191.6, 174.6, 172.7, 132.6, 132.1, 88.9,
86.5, 37.6, 37.5, 34.3, 33.6, 24.5, 23.3, 20.9,
19.0, 19.1, 18.9, 18.5, 12.5, 12.6, and 11.4.
258
-,cm
-m
-
-n
-W
(51
Z,
a.0
S
w-
p
~
259
(i-Pr)3Si
C
,.
+
250
NO
NO2
250
251
175
6-Nitro-3,4,6-trimethyl-2-triisopropylsilyl-2-cyclohexen-1-one
(251).
A flame-dried, threaded Pyrex (20 mm O.D., 16 mm I.D.) was charged with a
solution of silylketene 175 (0.154 g, 0.610 mmol), nitropropene (0.10 mL, 0.104 g, 1.20
mmol), and one crystal of BHT in 0.6 mL of toluene and then sealed with a rubber septum.
The reaction mixure was degassed (three freeze-pump-thaw cycles at -196 °C, < 0.5 mm
Hg) and tightly sealed with a teflon cap. The reaction mixture was heated at 110 °C for 16
h and then transferred to a round-bottomed flask using 1 mL of benzene and concentrated
to afford 0.23 g of a brown oil. Column chromatography on 15 g of silica gel (gradient
elution with 0-10% EtOAc-hexane) provided 0.073 g (35%) of 251 as a white solid: mp
82-84 °C.
IR (CC14):
2930, 2860, 1670, 1540, 1455, 1380, 1350, 1280,
and 1220 cm-1
1H
2.71 (dd, J = 13.5, 9.8 Hz,
2.27 (dd, J = 13.5, 5.8 Hz,
(s, 3H), 1.43 (sept, J = 7.4
7.1 Hz, 3H), and 1.04 (app
NMR (300 MHz, CDC13 ):
13C NMR (75 MHz, CDC13 ):
1H), 2.52 (m,
1H), 2.08 (s,
Hz, 3H), 1.28
t, J = 7.6 Hz,
1H),
3H), 1.78
(d, J =
18H)
194.9, 171.8, 132.2, 93.1, 40.4, 37.0, 23.4, 20.9,
20.4, 19.1, 19.0, 12.7, and 12.5
260
c
-r
-0
0
0
Cu
0
L
i
o
o
_ en
0
I-
uI
CN
c,
0
0o
00
0
0)
0
0D
261
ARQiILDrA_
1l
+
175
C
II
C
II
225
256
4,5,6-Trimethyl-2-[(triisopropylsilyl)oxy]-benzonitrile
(256).
A flame-dried, threaded Pyrex tube (20 mm O.D., 16 mm I.D.) was charged with a
solution of silylketene 175 (0.153 g, 0.605 mmol) and cyanoallene (0.132 g, 2.03 mmol)
in 0.6 mL of toluene and sealed with a rubber septum. The reaction mixure was degassed
(three freeze-pump-thaw
cycles at -196 °C, < 0.5 mm Hg) and tightly sealed with a teflon
cap. The reaction mixture was heated at 150 °C for 31 h and then transferred to a round-
bottomed flask using 1 mL of benzene and concentrated to afford a brown oil. Column
chromatography on 10 g of silica gel (gradient elution with 0-45% CH2C12-petroleum
ether) provided 0.13 g (67%) of 256 as white crystals: mp 58-59 °C.
IR (CC14):
2950, 2870, 2225, 1730, 1550, 1470, 1330, and
1240 cm - 1
1H
NMR (300 MHz, CDCl 3 ):
6.55 (s, 1H), 2.43 (s, 3H), 2.26 (s, 3H), 2.11 (s,
3H), 1.32 (sept, J = 6.9 Hz, 3H), and 1.13 (d, J =
7.0 Hz, 18H)
13C
NMR (75 MHz, CDC13 ):
IUV
max (CH 3 CN):
HRMS:
155.9, 142.5, 140.7, 127.9, 117.7, 117.0, 103.5,
21.6, 18.9, 18.0, 15.1, and 13.0
222 nm (e = 470)
Calcd for C19 H3 1ONSi:
Found:
262
317.2175
317.2165
O
o
.o
!
-o
C
o
cu
0woJ
0
0'p
o
"
IL,
_of
C4
0
L0
0
rr
_ -o
- 0o
o)
o
0
263
(i-Pr) 3 Si
C
+
(i-Pr) 3 Si
CH2N2
265
175
3,4-Dimethyl-2-triisopropylsilyl
cyclopent-2-en-1-one
(265).
A 10-mL, one-necked, round-bottomed flask (note: clearseal joint) was charged
with a solution of silylketene 175 (0.208 g, 0.82 mmol) in 4 mL of dichloromethane
and
cooled at -120 °C in a pentane-liquid nitrogen bath. A solution of CH 2 N2 (ca. 1.6 mmol)
generated from diazald (0.47 g, 2.2 mmol) was added rapidly to the silylketene solution
over 0.5 min via a fire-polished pipette. The reaction mixture was allowed to warm to
room temperature over 3 h and then concentrated to provide 0.24 g of a yellow oil.
Column chromatography on 10 g of silica gel (gradient elution with 0-10% ethyl acetatehexane) afforded 0.21 g (96%) of cyclopentenone 265 as a yellow oil..
IR (thin film):
2940, 2860, 1685, 1570, 1460, 1370, 1250, 1140,
1015, and 990 cm-1
1H
NMR (300 MHz, CDC13 ):
2.74 (app quintet, J = 7.0 Hz, 1H), 2.58 (dd, J =
18.3, 7.0 Hz, 1H), 2.15 (s, 3H), 1.96 (dd, J =
18.1, 2.2 Hz, 1H), 1.50 (sept, J = 7.1 Hz, 3H),
1.17 (d, J = 6.9 Hz, 3H), and 1.02 (d, J = 7.7 Hz,
18H)
13
C NMR (75 MHz, CDC13):
213.3, 191.1, 135.5, 44.7, 41.4, 20.0, 18.8, 17.7,
and 11.7
264
-o
.I
_o
0
i
. mmmomliw
I
I
-o
0
I1
m
o
-oo
tto
us
it-1
-r
o
o
.,Q
II
I
I
265
A
H
(;.Pr)3Si
(/P)3i
C
)CH
3
CH 3
"
175
265
3,4-Dimethyl-2-triisopropylsilyl
cyclopent-2-en-l-one
(265).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of trimethylsulfonium iodide (0.155 g,
0.762 mmol) in 1.7 mL of DME and cooled at 0 °C in an ice bath while n-butyllithium
solution (2.61 M in hexane, 0.25 mL, 0.65 mmol) was added over 0.5 min and the
resulting solution of ylide was stirred for 3 min at 0 °C. A 25-mL, two-necked, roundbottomed flask equipped with an argon inlet adapter and a rubber septum was charged with
a solution of silylketene 175 (0.156 g, 0.619 mmol) in 3.3 mL of DME and cooled at 0 °C.
The ylide solution was then transferred dropwise via cannula over 2 min to the ketene
solution, and the resulting mixture was stirred for 5 min at 0 °C. The cooling bath was then
removed, and the reaction mixture was allowed to warm to room temperature over 1.5 h.
The reaction mixture was diluted with 5 mL of water and extracted with three 3-mL
portions of diethyl ether. The combined organic phases were washed with 5 mL of
saturated NaCl solution, dried over Na2SO4, filtered, and concentrated to give 0.17 g of a
yellow oil. Column chromatography (twice) on 10 g of silica gel (gradient elution with 030% dichloromethane-petroleum
ether) afforded 0.10 g (61%) of cyclopentenone 265 as a
colorless oil with spectral characteristics identical with those listed above.
266
is-O
H <+
< ,CH
+H2C--SCH3
CH3
175
3,4-Dimethyl-2-triisopropylsilyl
265
cyclopent-2-en-1-one
(265).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of trimethylsulfonium iodide (0.118 g,
0.581 mmol) in 1.4 mL of THF and cooled at 0 °C in an ice bath while n-butyllithium
solution (2.50 M in hexane, 0.20 mL, 0.51 mmol) was added over 0.5 min and the
resulting solution of ylide was stirred for 3 min at 0 °C. A 25-mL, two-necked, roundbottomed flask equipped with an argon inlet adapter and a rubber septum was charged with
a solution of silylketene 175 (0.122 g, 0.482 mmol) in 2.4 mL of 50:50 THF-DMSO and
cooled at 0 °C. The ylide solution was then transferred dropwise via cannula over 2 min to
the ketene solution, and the resulting mixture was stirred for 30 min at 0 °C. The cooling
bath was then removed, and the reaction mixture was allowed to warm to room temperature
over 1 h. The reaction mixture was diluted with 5 mL of water and extracted with three 2mL portions of diethyl ether. The combined organic phases were washed with 5 mL of
saturated NaCl solution, dried over Na2SO4, filtered, and concentrated to give 0.13 g of a
yellow oil. Column chromatography (twice) on 10 g of silica gel (gradient elution with 05% ethyl acetate-hexane) afforded 0.097 g (75%) of cyclopentenone 265 as a colorless oil
with spectral characteristics identical with those listed above.
267
(i-Pr) 3 Si
C
H2C-- CH
',:.CH 3
175
3,4-Dimethyl-2-triisopropylsilyl
265
cyclopent-2-en-1-one (265).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with sodium hydride (0.023 g, 0.585 mmol) as a 60%
mineral oil dispersion which was washed with three 1-mL portions of hexane by swirling,
allowing the hydride to settle, and decanting in order to remove the mineral oil. The rubber
septum was replaced and the flask was fitted with a reflux condenser. A solution of
trimethylsulfoxonium
iodide (0.113 g, 0.515 mmol) in 0.5 mL of DMSO was added to the
dispersion and the resulting mixture was stirred for 15 min. A 25-mL, two-necked, round-
bottomed flask equipped with an argon inlet adapter and a rubber septum was charged with
a solution of silylketene 175 (0.100 g, 0.395 mmol) in 2 mL of a 1:1 DMSO-THF solution
and cooled at 0 °C in an ice bath. The ylide solution described above was then transferred
dropwise via cannula over 2 min to the ketene solution (the flask was rinsed with an
additional 0.5 mL of THF). The reaction mixture was stirred for 1.25 h and then diluted
with 5 mL of water and extracted with three 2-mL portions of diethyl ether. The combined
organic phases were washed with 5 mL of saturated NaCl solution, dried over Na2SO4,
and concentrated to give 0.12 g of a yellow oil. Column chromatography on 10 g of silica
gel (gradient elution with 0-5% ethyl acetate-hexane) afforded 0.045 g (43%) of
cyclopentenone 265 as a colorless oil with spectral characteristics identical to those listed
above.
268
C
(/Pr)3Si
0
/
(/Pr) 3 Si
+ CH(SiMe 3)N 2
SiMe3
277
175
trans-3,4-Dimethyl-2-triisopropylsilyl-5-trimethylsilyl
cyclopent-2-en-1-
one (277).
A 5-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of silylketene 175 (0.105 g, 0.41 mmol)
in 0.4 mL of dichloromethane. A solution of (trimethylsilyl)diazomethane (1.7 M in
hexane, 0.40 mL, 0.077 g, 0.68 mmol) was added dropwise over 30 sec, and the reaction
mixture was stirred at room temperature for 32 h. The resulting yellow solution was then
concentrated to afford 0.16 g of a yellow oil. Column chromatography on 10 g of silica gel
(gradient elution with 0-5% ethyl acetate-hexane) afforded 0.12 g (86%) of cyclopentenone
277 as a colorless oil.
IR (CC14 ):
2945, 2955, 1665, 1575, 1455, 1365, and
1245 cm-1
1H
NMR (300 MHz, CDC13 ):
(s,
2.56 (dq, J = 0.8, 7.4 Hz, 1H), 2.15 (s, 3H), 1.82
1H), 1.52 (sept, J = 7.4 Hz, 3H), 1.18 (d, J = 7.0
Hz, 3H), 1.04 (d, J = 7.5 Hz, 18H), and 0.05 (s,
9H)
13
C NMR (75 MHz, CDC13 ):
HRMS:
213.9, 188.9, 135.8, 50.9, 45.6, 21.3, 18.9, 18.7,
11.8, and -2.8
Calcd for C 1 9 H 3 8OSi 2 :
Found:
269
338.2461
338.2459
r
.0.
-o
-Cu
-m
-in
-c
0.L.
Q.
-a)
-o
b
O
270
-
Et3 Si
C'
+ CH(SiMe 3)N2
280
176
trans-3,4-Dimethyl-2-triethylsilyl-5-trimethylsilyl
cyclopent-2-en-l-one
(176).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of silylketene 176 (0.105 g, 0.500
mmol) in 0.5 mL of dichloromethane. A solution of (trimethylsilyl)diazomethane
(1.5 M in
hexane, 0.50 mL, 0.75 mmol) was added dropwise over 30 sec, and the reaction mixture
was stirred at room temperature for 21.5 h. The reaction mixture was then concentrated to
afford 0.16 g of a yellow oil. Column chromatography on 10 g of silica gel (gradient
elution with 0-5% ethyl acetate-hexane) afforded 0.14 g (95%) of cyclopentenone 280 as a
yellow oil.
IR (CC14):
2941, 2861, 1663, 1587, 1451, 1412, 1370, 1251,
1146, 1105, and 1001 cm - 1
1H
NMR (300 MHz, CDCl 3 ):
2.52 (qd, J = 6.2, 1.9 Hz, 1H), 2.05 (s, 3H), 1.73
(d, J = 1.8 Hz, 1H), 1.11 (d, J = 6.7 Hz, 3H), 0.86
(t, J = 6.9 Hz, 9H), 0.71 (q, J = 6.9 Hz, 6H), and
-0.01 (s, 9H)
13
C NMR (75 MHz, CDC13):
HRMS:
188.2, 184.9, 137.1, 50.6, 45.3, 20.7, 17.6, 7.4,
3.5, and -3.0
Calcd for C 16 H 3 2OSi 2 :
Found:
271
296.1992
296.1988
O
0
- mO
X
-0
0
rm
~U)i~~
o
272
t
0
-0
i
0o
II0
C
_0,
272
(I-Pr) 3 SI
C._
+
SiMe3
(i-Pr) 3Si,
CH(SiMe 3)N 2
Ph
281
182
trans-4-phenyl-2-tri isopropylsilyl-5-trimethylsilylcyclopent-2-en- 1-one
(281).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of silylketene 182 (0.130 g, 0.433
mmol) in 0.4 mL of dichloromethane.
A solution of (trimethylsilyl)diazomethane
(1.2 M in
hexane, 0.55 mL, 0.65 mmol) was added dropwise over 30 sec and the resulting mixture
was stirred at room temperature for 22 h. The reaction mixture was concentrated to afford
0.15 g of a yellow oil. Column chromatography on 10 g of silica gel (gradient elution with
0-2% ethyl acetate-hexane) afforded 0.14 g (81%) of cyclopentenone 281 as a yellow oil.
IR (CC14):
3030, 2959, 2869, 1770, 1678, 1570, 1490, 1460,
1383, 1280, 1250, 1160, 1070, 1040, and 1020
cm- 1
:H NMR (300 MHz, CDC13):
7.69 (d, J = 2.8 Hz, 1H), 7.12-7.30 (m, 3H), 7.10
(d, J = 7.6 Hz, 2H), 4.00 (s, 1H), 2.19 (d, J = 2.1
Hz, 1H), 1.36 (sept, J = 7.7 Hz, 3H), 1.09 (d, J
= 8.2 Hz, 6H), 1.06 (d, J = 7.9 Hz, 12 H), and
0.13 (s, 9H)
13 C NMR (75 MHz, CDC13):
214.8, 174.6, 143.0, 142.3, 129.0, 127.3, 127.2,
126.9, 52.7, 51.7, 18.7, 11.0, and -2.6
HRMS:
Calcd for C23 H 3 8OSi2 :
Found:
273
386.2461
386.2457
I
aI
0
0
'I I
mlv
I
_
:
Cu
0c
m
I
71
0
1
0r
-- v
A
,-1
i
4i
L
I-0
-.
274
0
(i-Pr 3 )Si
(i-Pr3
C
+
SiMe 3
CH(SiMe 3)N 2
282
174
7-Trimethylsilyl-9- (triisopropylsilyl)bicyclo[4.3.0]non-1
(9)-en-8-one
(282).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of silylketene 174 (0.119 g, 0.430
mmol) in 0.4 mL of dichloromethane.
A solution of (trimethylsilyl)diazomethane
(1.7 M in
hexane, 0.38 mL, 0.64 mmol) was added dropwise over 30 sec and the reaction mixture
was stirred at room temperature for 29.5 h. The reaction mixture was concentrated to
afford 0.200 g of a yellow oil. Column chromatography on 10 g of silica gel (gradient
elution with 0-5% ethyl acetate-hexane) afforded 0.130 g (83%) of cyclopentenone 282 as
a colorless oil.
IR (CCh):
2940, 2870, 1685, 1575, 1460, 1440, 1350, and
1290 cm-1
1H NMR (300 MHz, CDC13):
2.96 (m, 1H), 2.40 (ddd, J = 12.0, 5.0, 2.0 Hz,
1H), 2.23 (dd, J = 13.3, 5.3 Hz, 1H), 1.76-2.17
(m, 2H), 1.73 (d, J = 1.0 Hz,lH), 1.52 (sept, J =
7.7 Hz, 3H), 1.02 (d, J = 7.4 Hz, 18H), 0.80-1.60
(m, 4H), and 0.05 (s, 9H)
13 C NMR (75 MHz, CDC13):
212.0, 190.9, 132.8, 48.3, 48.0, 37.1, 32.4, 28.1,
25.6, 18.9, 11.7, and -2.7
HRMS:
Calcd for C 2 lH 4 0OSi2 :
Found:
275
364.2617
364.2616
x
0.
CL
-Mn
S
E
X N
co
-
v
-m
-Io
276
(i-Pr)3Si
Co.
0
H2C--SH
+
3
(i-Pr)
3Si
CH3
Ph
Ph
182
296
4-Phenyl-2-triisopropylsilylcyclopenten-1-one
(296).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of trimethylsulfonium iodide (0.0891 g,
0.437 mmol) in 1.1 mL of DME and cooled at 0 °C in an ice bath while n-butyllithium
solution (2.65 M in hexane, 0.16 mL, 0.421 mmol) was added over 0.5 min and the
resulting ylide was stirred for 5 min at 0 °C. A 25-mL, two-necked, round-bottomed flask
equipped with an argon inlet adapter and a rubber septum was charged with a solution of
silylketene 182 (0.121 g, 0.401 mmol) in 1.8 mL of DME and cooled at 0 °C. The ylide
solution was then transferred dropwise via cannula over 2 min to the ketene solution (the
flask was rinsed with 0.5 mL of additional DME). The cooling bath was then removed and
the reaction mixture was allowed to warm to room temperature for 1 h and then diluted with
5 mL of water. The aqueous layer was separated extracted with three 3-mL portions of
diethyl ether, and the combined organic phases were washed with 5 mL of saturated NaCl
solution, dried over Na2SO4, filtered, and concentrated to give 0.17 g of a yellow oil.
Column chromatography on 10 g of silica gel (gradient elution with 0-5% ethyl acetatehexane) afforded 0.082 g (65%) of cyclopentenone 296 as a colorless oil.
IR (CC14):
2940, 2860, 1695, 1565, 1490, 1460, 1410, 1380,
1280, 1260, and 1060 cm-1
1H
7.78 (d, J = 2.4 Hz, 1H), 7.23-7.37 (m, 3H), 7.13
NMR (300 MHz, CDC13):
(d, J = 6.9 Hz, 2H), 4.17 (dt, J = 7.5, 2.3 Hz, 1H),
2.90 (dd, J = 18.8, 7.0 Hz, 1H), 2.33 (dd, J =
18.4, 3.0 Hz, 1H), 1.37 (sept, J = 7.3 Hz, 3H),
277
1.08 (d, J = 7.2 Hz, 12H), and 1.07 (d, J = 7.2 Hz,
6H)
13 C NMR (75 MHz, CDC13):
213.6, 176.2, 142.6, 142.2, 129.0, 127.2, 127.0,
48.4, 45.1, 18.6, and 10.9
278
-a.
-o
---
122
-'
-0
-O
0o
L
CL
a-
o =9'
m
-o
CL
tr
Li\
t
o
-D
-o
279
-B
o
-
-
Et3 Si
C
D )
+
CH3
H2 C-S\
CH 3
297
176
3,4-Dimethyl-2-triethylsilyl
cyclopent-2-en-l-one
(297).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of trimethylsulfonium iodide (0.091 g,
0.448 mmol) in 1.1 mL of tetrahydrofuran
and cooled at 0 C in an ice bath while n-
butyllithium (2.48 M in hexane, 0.15 mL, 0.372 mmol) was added over 0.5 min and the
resulting ylide was stirred for 2 min at 0 °C. A 25-mL, two-necked, round-bottomed flask
equipped with an argon inlet adapter and a rubber septum was charged with a solution of
silylketene 176 (0.0764 g, 0.365 mmol) in 1.8 mL of 50:50 THF-DMSO and cooled at 0
°C. The ylide solution was then transferred dropwise via cannula over 2 min to the ketene
solution (the flask was rinsed with 0.5 mL of additional THF), and the reaction mixture
was stirred for 15 min at 0 "C. The cooling bath was removed and the reaction mixture was
allowed to warm to room temperature over 2 h and then diluted with 5 mL of water. The
aqueous layer was separated and extracted with five 3-mL portions of diethyl ether, and the
combined organic phases were washed with 5 mL of saturated NaCl solution, dried over
Na 2 SO4, and concentrated to give 0.083 g of a yellow oil. Column chromatography on 10
g of silica gel (gradient elution with 0-5% ethyl acetate-hexane) afforded 0.062 g (75%) of
cyclopentenone 297 as a colorless oil.
IR (CC14):
2940, 2885, 1685, 1580, 1410, 1250, 1200, and
1065 cm ' 1
1H
2.78 (app quintet, J = 7.1 Hz, 1H), 2.59 (dd, J =
18.6, 7.0 Hz, 1H), 2.13 (s, 3H), 1.97 (dd, J =
18.3, 2.4 Hz, 1H), 1.18 (d, J = 7.1 Hz, 3H), 0.97
NMR (300 MHz, CDCl 3):
280
(t, J = 7.6 Hz, 3H), 0.91 (t, J = 6.9 Hz, 6H),
0.78 (q, J = 7.5 Hz, 2H), and 0.77 (t, J = 6.7 Hz,
4H)
13 C NMR (75 MHz, CDC13):
213.2, 190.6, 136.6, 44.6, 41.1, 19.5, 18.0, 7.4,
HRMS:
Calcd for C13 H24OSi:
and 3.4
Found:
281
224.1596
224.1594
- X
- X.
---
e
--
-
In
is
w
-(D
In
-T
-<D
-M
282
(i-Pr) 3 SI
C' '
0
e
®/Ph
(IPr)Si
3
+ CH3 CH-S
(Pr)3
Ph
175
298
trans-3,4,5-Trimethyl-2-triisopropylsilyl
cyclopent-2-en-1-one (298).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of diphenylethylsulphonium fluoroborate
(0.087 g, 0.287 mmol) in 1.1 mL of DME and cooled at -50 °C in an dry ice-acetone bath
whilet-butyllithium (1.84 M in pentane, 0.15 mL, 0.281 mmol) was added over 0.5 min.
The resulting bright yellow ylide solution was stirred for 30 min at -50 °C. A 25-mL, two-
necked, round-bottomed flask equipped with an argon inlet adapter and a rubber septum
was charged with a solution of silylketene 175 (0.0676 g, 0.268 mmol) in 1.1 mL of DME
and cooled at -50 C. The ylide solution was then transferred dropwise via cannula over 2
min to the ketene solution (the flask was rinsed with an additional 0.5 mL of DME). The
cooling bath was removed, and the reaction mixture was allowed to warm from -50 °C to
-20 C over 1 h, and then diluted with 5 mL of water. The aqueous layer was separated
and extracted with three 2-mL portions of diethyl ether, and the combined organic phases
were washed with 5 mL of saturated NaCl solution, dried over Na2SO4, and concentrated
to give 0.092 g of a yellow oil. Column chromatography (twice) on 10 g of silica gel
(gradient elution with 0-5% ethyl acetate-hexane)
afforded 0.051 g (68%) of
cyclopentenone 298 as a colorless oil.
IR (CC14):
2940, 2882, 1685, 1570, 1455, 1370, 1255, 1225,
1142, 1020, 990, and 890 cm -1
li NMR (300 MHz, CDC13):
2.34 (qd, J = 6.9, 3.5 Hz, 1H), 2.15 (s, 3H), 1.92
(qd, J = 7.3, 3.2 Hz, 1H), 1.52 (sept, J = 7.2 Hz,
3H), 1.21 (d, J = 6.6 Hz, 3H), 1.13 (d, J = 7.3 Hz,
283
3H), 1.05 (d, J = 7.2 Hz, 12H), and 1.04 (d, J =
8.0 Hz, 6H)
13 C
NMR (75 MHz, CDC13):
215.2, 188.7, 134.1, 50.1, 49.9, 18.9, 18.8, 18.7,
15.5, and 11.8
284
I
.
Q
o
cn
o
0
v
Cr.
A.4
IA
o
-(
285
0
O
a
265
3,4-Dimethyl
cyclopenten-l-one
316
(316).
A 10-mL, two-necked, round-bottomed flask equipped with an argon inlet adapter
and a rubber septum was charged with a solution of silylcyclopentenone 265 (0.114 g,
0.426 mmol) in 3.5 mL of methanol. Methanesulfonic acid (0.45 mL, 0.66 g, 6.90 mmol)
was added rapidly dropwise, and the resulting mixture was stirred at room temperature for
3 h. The reaction mixture was diluted with 5 mL of dichloromethane and transferred to a
separatory funnel. The aqueous phase was extracted with two 2-mL portions of CH2C12,
and the combined organic layers were washed with 5 mL of saturated NaCl solution, dried
over Na2SO4, filtered, and concentrated to afford 0.094 g of a colorless oil. Column
chromatography on 10 g of silica gel (gradient elution with 0-30% ethyl acetate-hexane)
afforded 0.045 g (95%) of cyclopentenone 316172 as a colorless oil.
IR (CC14):
2920, 2860, 1710, 1620, 1540, 1430, 1370, 1260,
and 1000 cm-1
1H
5.89 (s, 1H), 2.79 (app quintet, J = 7.0 Hz, 1H),
2.64 (dd, J = 18.6, 6.7 Hz, 1H), 2.08 (s, 3H), 1.98
(dd, J = 18.0, 1.9 Hz, 1H), and 1.18 (d, J = 7.4,
3H)
NMR (300 MHz, CDC13):
13 C NMR (75 MHz, CDC13):
208.9, 182.8, 130.3, 53.4, 44.2, 18.8, and 17.1
286
1C -
-ll -
Cu
·
--
i
0
0
t
i
0
0
tl
0
0~~c
-0 o
.
-o
-
287
c,
_ D
0
Et3 Si
297
3,4-Dimethyl
316
cyclopent-2-en-1-one
(316).
A 25-mL, two-necked, pear-shaped flask equipped with an argon inlet adapter and
a rubber septum was charged with a solution of silylcyclopentenone 297 (0.131 g, 0.584
mmol) in 5.8 mL of methanol. Methanesulfonic acid (0.19 mL, 0.28 g, 2.92 mmol) was
added rapidly dropwise, and the reaction mixture was stirred at room temperature. After 4
h, a second portion of methanesulfonic acid (0.19 mL, 0.28 g, 2.92 mmol) was added, and
the reaction mixture was stirred for an additional 4.5 h and then diluted with 5 mL of water.
The aqueous phase was separated and extracted with three 5-mL portions of CH2C12 , and
the combined organic layers were washed with 15 mL of saturated NaCl solution, dried
over Na2SO4 , filtered, and concentrated to afford 0.12 g of a yellow-brown oil. Column
chromatography on 10 g of silica gel (gradient elution with 0-30% ethyl acetate-hexane)
afforded 0.057 g (89%) of cyclopentenone 316 as a colorless oil with spectral
characteristics identical to those listed above.
288